Citation for published version (APA): Yu, L. (2012). The role of galectin-3 in cardiac remodeling and fibrogenesis Groningen: s.n.

Transcription

1 University of Groningen The role of galectin-3 in cardiac remodeling and fibrogenesis Yu, Lili IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2012 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Yu, L. (2012). The role of galectin-3 in cardiac remodeling and fibrogenesis Groningen: s.n. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date:

2 The Role of Galectin-3 in Cardiac Remodeling and Fibrogenesis Lili Yu

3 Financial support by the Netherlands Heart Foundation for the publication of this thesis is gratefully acknowledged. Financial support by the Groningen University Institute for Drug Exploration (GUIDE) and the University of Groningen for the publication of this thesis is gratefully acknowledged. CIP-gegevens koninklijke bibiotheek, Den Haag Lili Yu The Role of Galectin-3 in Cardiac Remodeling and Fibrogenesis Proefschrift Groningen ISBN: ISBN: (digital version) Copyright 2012 Lili Yu All rights are reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without permission of the author. Cover information: Galectin-3 is combined with Chinese dragon totem Cover design: Lili Yu Lay-out: Cheng Qian and Lili Yu Printed by: Gildeprint Drukkerijen B.V., Enschede

4 The Role of Galectin-3 in Cardiac Remodeling and Fibrogenesis Proefschrift ter verkrijging van het doctoraat in de Medische Wetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. E. Sterken, in het openbaar te verdedigen op dinsdag 11 december 2012 om 9:00 uur door Lili Yu geboren op 28 juni 1976 te Heilongjiang, China

5 Promotor : Copromotores : Beoordelingscommissie: Prof. dr. W.H. van Gilst Dr. R.A.de Boer Dr. H.H.W.Silljé Prof. dr. A.A.Voors Prof. dr. H. van Goor Prof. dr. R.A.Bank

6 Paranimfen: Megan V. Cannon Hongjuan Yu We thank BG Medicine, Inc. (Waltham, MA, USA) for providing the financial support to print this thesis. Financial support by Novartis Pharma B.V. is gratefully acknowledged. Financial support by Bayer Health Care B.V. is gratefully acknowledged. Financial support by Servier Nederland Farma B.V. is gratefully acknowledged.

7

8 CONTENTS Chapter 1 Introduction and Aim of the Thesis 10 Chapter 2 Galectin-3 in Cardiac Remodeling and Heart Failure 21 Chapter 3 Role of Galectin-3 Pathway in the Pathogenesis of Cardiac Remodeling 37 and Heart Failure Chapter 4 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac 55 Remodeling by Interfering with Myocardial Fibrogenesis Chapter 5 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive 95 Nephropathy Chapter 6 Clinical Correlations of Plasma Galectin-3 Levels in a Well-Defined 113 Chronic Heart Failure Cohort Chapter 7 Summary and Perspectives 127 Nederlandse Samenvatting 139 Acknowledgements 149

9

10 Chapter 1 Introduction and Aim of the Thesis

11 Chapter 1 Introduction Chronic heart failure (CHF) is the leading cause of hospitalization in people older than age 65 [1]. In developing countries, 2-3 % of the population suffers from heart failure, but in those years old, it occurs in %. CHF is associated with significant morbidity and mortality attributable largely to adverse structural changes to the myocardium with related cardiac dysfunction (pump failure), arrhythmias and premature sudden death, and other associated complications such as renal disease. Cardiac remodeling and fibrogenesis Cardiac remodeling involves changes in size, shape, structure and physiology of the heart in response to elevated hemodynamic load and/or cardiac injury in association with neurohormonal activation [2]. Remodeling may be described physiologically or pathologically [3]. Physiological remodeling has been observed in athletes and has been termed athlete s heart, whereas pathological remodeling may occur with pressure overload (e.g. aortic stenosis, hypertension), volume overload (e.g. valvular stenosis or regurgitation), or following cardiac injury (e.g. myocardial infarction, myocarditis or idiopathic dilated cardiomyopathy). In the latter, capillary network density is less and therefore not capable of supplying the greater demand of the hypertrophied myocardium. [4-6]. In each of these settings, remodeling may transition from an apparently compensatory process to a maladaptive one [7]. The transition is specifically associated with changes to both the cellular and extracellular matrix (ECM) such as myocyte hypertrophy, apoptosis or necrosis, as well as fibroblast proliferation, myofibroblast activation and development of fibrosis. All these changes are influenced by various factors including increased hemodynamic load, neurohumoral activation via endothelin and cytokines signaling, oxidative stress, matrix metalloproteinases (MMPs), and inflammatory responses involving recruitment of peripheral monocytes and macrophages (Figure 1). Cardiac fibroblasts (CFB) are the most abundant cell type within the myocardium [8], which is critical in maintaining normal cardiac structure and ECM homeostasis [9-12]. In pathological remodeling, up-regulated CFB can result in excess matrix production and deposition of ECM proteins in the myocardium, such as collagen synthesis and deposition, which exerts adverse effects on cardiac structure and function [12-16]. In addition to being the primary source of ECM proteins, fibroblasts produce a number of cytokines, peptides and enzymes among which MMPs and tissue inhibitor of metalloproteinase (TIMPs) directly impact the ECM turnover and homeostasis [13-19]. Remodeling of the ECM plays a pivotal role in cardiac remodeling and is a key process determining the clinical course and outcome of cardiovascular diseases evolving with CHF. Recent research on the assessment and the tre- 10

12 Introduction and Aim of the Thesis 1 Figure 1. Pathological cardiac remodeling and fibrogenesis. -atment of CHF patients has expanded from an initial focus on reducing hemodynamic load to interventions that potentially modify maladaptive cellular and molecular processes. Accumulating evidence shows that, in the pathophysiological process of CHF, molecular biomarkers could provide a unique avenue to potentially improve our capability of predicting adverse outcomes, to serve as novel drug targets, and moreover, help gauge therapeutic efficacy. Some 'traditional' biomarkers such as cardiac troponin, natriuretic peptides, and C- reactive protein have been studied in the patients with CHF and are currently well-established parameters in clinical practice, respectively indicating the level of cardiac myocyte damage, inflammation, ventricular remodeling, myocardial injury and renal dysfunction that occurs in CHF [20, 21]. One such emerging biomarker is Galectin-3 (Gal-3), associated mainly with myocardial fibrogenesis, and which is properly studied in the present thesis. The characteristics of Gal-3 Gal-3 is a kda chimaera-type galectin belonging to the β-galactosidase binding lectins family, and is characterized by an extended N-terminal domain ( amino acids) and a C-terminal carbohydrate recognition domain (CRD) (130 amino acids) [22, 23]. Gal-3 communicates with a variety of ECM proteins, carbohydrates (e.g., N-acetyllactosamine) as well as unglycosylated proteins such as cell surface receptors (macrophage antigens CD11b/ CD18) and extracellular receptors (collagen IV) which, furthermore, modulates cell-cell adhe- 11

13 Chapter 1 Figure 2. Glactin-3 is expressed in various cells and organs. -sion signaling in the extracellular compartment [24-27]. Gal-3 is expressed in various types of cells and tissues and is important in diverse physiological and pathological processes such as immune and inflammatory responses, tumor development and progression, neural degeneration, atherosclerosis, diabetes, as well as wound repair. Gal-3 has been detected in many proliferative cells including tumor cells, eosinophils, neutrophils, activated macrophages and fibroblasts [28-31] (Figure 2). Notably, many of these cell types play active roles in the inflammatory response and the formation of fibrosis (fibrogenesis). Recent studies reveal that, as a multifunctional biomarker, Gal-3 plays a key role in the cardiac remodeling process by participating in ECM homeostasis and inflammation responses. [20] Gal-3 and its therapeutic implications Gal-3 was discovered around two decades ago [32]. It is widely distributed in several organs including the heart, lungs, liver and kidneys [33]. The role of Gal-3 in fibrosis and inflammation has been elucidated in recent years. In the healthy tissue, Gal-3 expression is absent or reduced. However, under pathological conditions, Gal-3 expression is substantially up-regulated, specifically in inflammation and fibrosis which are crucial in cardiac remodeling and renal fibrosis. In-situ hybridization and immunohistochemistry analyses show that Gal-3 is highly distributed in the fibrotic area of myocardium and co-localizes with macrophages 12

14 Introduction and Aim of the Thesis [34]. An ever-growing body of experimental evidence has found that macrophage-derived Gal-3 was associated with activated myofibroblasts and subsequently increased collagen synthesis and deposition, playing an important role in regulating extracellar matrix in the damaged tissue area [30, 31, 34-41]. The first evidence showing an involvement of Gal-3 in heart failure originates from a landmark study by Sharma and colleagues [34]. The researchers demonstrated Gal-3 as a new target for intervention in the CHF. Furthermore, Henderson et al demonstrated that Gal-3 expression was markedly increased in progressive renal fibrosis. Gal-3 deficiency exhibited less renal inflammation, representing a lesser pro-fibrotic response with significant decreases in collagen production and deposition [30]. Additionally, Kalatjou et al. showed significantly increased Gal-3 expression in a kidney fibrosis model, which was amenable to a novel therapeutic strategy, Gal-3 inhibition, to attenuate fibrosis [42]. In addition to experimental studies, various clinical trials have also indicated a potential clinical utility of Gal-3 as a biomarker for prognosticating heart failure. Van Kimmenade et al firstly evaluated the prognostic and predictive value of Gal-3 as a biomarker in acute heart failure [43]. In PREVEND trial, they then demonstrated that high plasma Gal-3 levels were associated with all-cause mortality [44]. Furthermore, the DEAL-HF trial investigated the incremental value of Gal-3 over Nt-proBNP alone [45]. Finally, the HF-ACTION study revealed that Gal-3 was associated, yet not independent from NT-proBNP, with NYHA class II-IV, lower systolic blood pressure, increased creatinin, and lower maximal oxygen consumption [46]. In summary, macrophage-derived Gal-3 is associated with myofibroblast-induced collagen synthesis and deposition, playing a central role in pathophysiological cardiac remodeling and heart failure. 1 The aim of this thesis As a multifunctional biomarker, Gal-3 plays an important role in detecting fibrogenesis and inflammatory processes, and is attracting widespread attention. Numerous studies have been elucidating the role of Gal-3 in the development of organ fibrosis, however, Gal-3 targeted therapy in cardiac remodeling or renal dysfunction that occurs in CHF has been given less attention. This thesis will discuss the role of Gal-3 in fibrogenesis of cardiac remodeling and cardiac-related kidney disease. The main focus will be on Gal-3-related fibrogenesis in cardiovascular diseases. Chapter 2 and 3 of this thesis will describe the mapping of Gal-3 pathways in the pathogenesis of cardiac remodeling in CHF and illustrate its potential therapeutic target in fibrogenesis by inhibiting collagen synthesis. The main question discussed in the above two chapters is: how does Gal-3 participate in cardiac remodeling as well as further influence heart failure progression. 13

15 Chapter 1 To further address Gal-3 targeting as a potential therapeutic candidate in cardiac remodeling, chapter 4 focuses on whether genetically or pharmacologically inhibiting Gal-3 can prevent or reverse cardiac remodeling. We will show how Gal-3 influences fibrogenesis by using different cardiac fibrosis models. Furthermore, since cardiovascular diseases are frequently accompanied with renal dysfunction, we will subsequently investigate Gal-3 targeted intervention in a hypertensive rat model (chapter 5). Moreover, we will also discuss the correlations that exist between the plasma Gal-3 levels and renal dysfunction that occurs in a well-defined clinical CHF cohort (chapter 6). 14

16 Introduction and Aim of the Thesis References 1. Krumholz HM, Chen YT, Wang Y, Vaccarino V, Radford MJ, Horwitz RI (2000) Predictors of readmission among elderly survivors of admission with heart failure. Am Heart J 139: Mihl C, Dassen WR, Kuipers H (2008) Cardiac remodelling: concentric versus eccentric hypertrophy in strength and endurance athletes. Neth Heart J 16: Cohn JN, Ferrari R, Sharpe N (2000) Cardiac remodeling--concepts and clinical implications: a consensus paper from an international forum on cardiac remodeling. Behalf of an International Forum on Cardiac Remodeling. J Am Coll Cardiol 35: de Boer RA, Pinto YM, Suurmeijer AJ, Pokharel S, Scholtens E, Humler M, Saavedra JM, Boomsma F, van Gilst WH, van Veldhuisen DJ (2003) Increased expression of cardiac angiotensin II type 1 (AT(1)) receptors decreases myocardial microvessel density after experimental myocardial infarction. Cardiovasc Res 57: De Boer RA, Pinto YM, Van Veldhuisen DJ (2003) The imbalance between oxygen demand and supply as a potential mechanism in the pathophysiology of heart failure: the role of microvascular growth and abnormalities. Microcirculation 10: Sano M, Minamino T, Toko H, Miyauchi H, Orimo M, Qin Y, Akazawa H, Tateno K, Kayama Y, Harada M, Shimizu I, Asahara T, Hamada H, Tomita S, Molkentin JD, Zou Y, Komuro I (2007) p53-induced inhibition of Hif-1 causes cardiac dysfunction during pressure overload. Nature 446: Opie LH, Commerford PJ, Gersh BJ, Pfeffer MA (2006) Controversies in ventricular remodelling. Lancet 367: Camelliti P, Borg TK, Kohl P (2005) Structural and functional characterisation of cardiac fibroblasts. Cardiovasc Res 65: Camelliti P, Green CR, LeGrice I, Kohl P (2004) Fibroblast network in rabbit sinoatrial node: structural and functional identification of homogeneous and heterogeneous cell coupling. Circ Res 94: Kohl P (2003) Heterogeneous cell coupling in the heart: an electrophysiological role for fibroblasts. Circ Res 93: Gaudesius G, Miragoli M, Thomas SP, Rohr S (2003) Coupling of cardiac electrical activity over extended distances by fibroblasts of cardiac origin. Circ Res 93: Porter KE, Turner NA (2009) Cardiac fibroblasts: at the heart of myocardial remodeling. Pharmacol Ther 123: Eghbali M (1992) Cardiac fibroblasts: function, regulation of gene expression, and phenotypic modulation. Basic Res Cardiol 87 Suppl 2: Butt RP, Laurent GJ, Bishop JE (1995) Collagen production and replication by cardiac fibroblasts is enhanced in response to diverse classes of growth factors. Eur J Cell Biol 68: Diez J, Querejeta R, Lopez B, Gonzalez A, Larman M, Martinez Ubago JL (2002) Losartan-dependent regression of myocardial fibrosis is associated with reduction of left ventricular chamber stiffness in hypertensive patients. Circulation 105: Zannad F, Rossignol P, Iraqi W (2010) Extracellular matrix fibrotic markers in heart failure. Heart Fail Rev 15:

17 Chapter Moore L, Fan D, Basu R, Kandalam V, Kassiri Z (2012) Tissue inhibitor of metalloproteinases (TIMPs) in heart failure. Heart Fail Rev 17: Spinale FG (2007) Myocardial matrix remodeling and the matrix metalloproteinases: influence on cardiac form and function. Physiol Rev 87: Spinale FG, Wilbur NM (2009) Matrix metalloproteinase therapy in heart failure. Curr Treat Options Cardiovasc Med 11: Ahmad T, Fiuzat M, Felker GM, O'Connor C (2012) Novel biomarkers in chronic heart failure. Nat Rev Cardiol 9: Smilde TD, Damman K, van der Harst P, Navis G, Westenbrink BD, Voors AA, Boomsma F, van Veldhuisen DJ, Hillege HL (2009) Differential associations between renal function and "modifiable" risk factors in patients with chronic heart failure. Clin Res Cardiol 98: Seetharaman J, Kanigsberg A, Slaaby R, Leffler H, Barondes SH, Rini JM (1998) X-ray crystal structure of the human galectin-3 carbohydrate recognition domain at 2.1-A resolution. J Biol Chem 273: Henrick K, Bawumia S, Barboni EA, Mehul B, Hughes RC (1998) Evidence for subsites in the galectins involved in sugar binding at the nonreducing end of the central galactose of oligosaccharide ligands: sequence analysis, homology modeling and mutagenesis studies of hamster galectin-3. Glycobiology 8: Rosenberg I, Cherayil BJ, Isselbacher KJ, Pillai S (1991) Mac-2-binding glycoproteins. Putative ligands for a cytosolic beta-galactoside lectin. J Biol Chem 266: Sato S, Hughes RC (1992) Binding specificity of a baby hamster kidney lectin for H type I and II chains, polylactosamine glycans, and appropriately glycosylated forms of laminin and fibronectin. J Biol Chem 267: Ochieng J, Furtak V, Lukyanov P (2004) Extracellular functions of galectin-3. Glycoconj J 19: de Boer RA, Yu L, van Veldhuisen DJ (2010) Galectin-3 in cardiac remodeling and heart failure. Curr Heart Fail Rep 7: Hughes RC (1997) The galectin family of mammalian carbohydrate-binding molecules. Biochem Soc Trans 25: Hughes RC (1999) Secretion of the galectin family of mammalian carbohydrate-binding proteins. Biochim Biophys Acta 1473: Henderson NC, Mackinnon AC, Farnworth SL, Kipari T, Haslett C, Iredale JP, Liu FT, Hughes J, Sethi T (2008) Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol 172: Henderson NC, Mackinnon AC, Farnworth SL, Poirier F, Russo FP, Iredale JP, Haslett C, Simpson KJ, Sethi T (2006) Galectin-3 regulates myofibroblast activation and hepatic fibrosis. Proc Natl Acad Sci U S A 103: Cherayil BJ, Chaitovitz S, Wong C, Pillai S (1990) Molecular cloning of a human macrophage lectin specific for galactose. Proc Natl Acad Sci U S A 87: Yang RY, Rabinovich GA, Liu FT (2008) Galectins: structure, function and therapeutic potential. Expert Rev Mol Med 10:e17 16

18 Introduction and Aim of the Thesis 34. Sharma UC, Pokharel S, van Brakel TJ, van Berlo JH, Cleutjens JP, Schroen B, Andre S, Crijns HJ, Gabius HJ, Maessen J, Pinto YM (2004) Galectin-3 marks activated macrophages in failure-prone hypertrophied hearts and contributes to cardiac dysfunction. Circulation 110: Kasper M, Hughes RC (1996) Immunocytochemical evidence for a modulation of galectin 3 (Mac-2), a carbohydrate binding protein, in pulmonary fibrosis. J Pathol 179: Wang L, Friess H, Zhu Z, Frigeri L, Zimmermann A, Korc M, Berberat PO, Buchler MW (2000) Galectin-1 and galectin- 3 in chronic pancreatitis. Lab Invest 80: Sasaki S, Bao Q, Hughes RC (1999) Galectin-3 modulates rat mesangial cell proliferation and matrix synthesis during experimental glomerulonephritis induced by anti-thy1.1 antibodies. J Pathol 187: Eis V, Luckow B, Vielhauer V, Siveke JT, Linde Y, Segerer S, Perez De Lema G, Cohen CD, Kretzler M, Mack M, Horuk R, Murphy PM, Gao JL, Hudkins KL, Alpers CE, Grone HJ, Schlondorff D, Anders HJ (2004) Chemokine receptor CCR1 but not CCR5 mediates leukocyte recruitment and subsequent renal fibrosis after unilateral ureteral obstruction. J Am Soc Nephrol 15: Sharma U, Rhaleb NE, Pokharel S, Harding P, Rasoul S, Peng H, Carretero OA (2008) Novel anti-inflammatory mechanisms of N-Acetyl-Ser-Asp-Lys-Pro in hypertension-induced target organ damage. Am J Physiol Heart Circ Physiol 294:H Liu YH, D'Ambrosio M, Liao TD, Peng H, Rhaleb NE, Sharma U, Andre S, Gabius HJ, Carretero OA (2009) N-acetylseryl-aspartyl-lysyl-proline prevents cardiac remodeling and dysfunction induced by galectin-3, a mammalian adhesion/growth-regulatory lectin. Am J Physiol Heart Circ Physiol 296:H Mackinnon AC, Gibbons MA, Farnworth SL, Leffler H, Nilsson UJ, Delaine T, Simpson AJ, Forbes SJ, Hirani N, Gauldie J, Sethi T (2012) Regulation of Transforming Growth Factor-beta1-driven Lung Fibrosis by Galectin-3. Am J Respir Crit Care Med 185: Kolatsi-Joannou M, Price KL, Winyard PJ, Long DA (2011) Modified citrus pectin reduces galectin-3 expression and disease severity in experimental acute kidney injury. PLoS One 6:e van Kimmenade RR, Januzzi JL,Jr, Ellinor PT, Sharma UC, Bakker JA, Low AF, Martinez A, Crijns HJ, MacRae CA, Menheere PP, Pinto YM (2006) Utility of amino-terminal pro-brain natriuretic peptide, galectin-3, and apelin for the evaluation of patients with acute heart failure. J Am Coll Cardiol 48: de Boer RA, van Veldhuisen DJ, Gansevoort RT, Muller Kobold AC, van Gilst WH, Hillege HL, Bakker SJ, van der Harst P (2011) The fibrosis marker galectin-3 and outcome in the general population. J Intern Med 45. Lok DJ, Van Der Meer P, de la Porte PW, Lipsic E, Van Wijngaarden J, Hillege HL, van Veldhuisen DJ (2010) Prognostic value of galectin-3, a novel marker of fibrosis, in patients with chronic heart failure: data from the DEAL-HF study. Clin Res Cardiol 99: Felker GM, Fiuzat M, Shaw LK, Clare R, Whellan DJ, Bettari L, Shirolkar SC, Donahue M, Kitzman DW, Zannad F, Pina IL, O'Connor CM (2012) Galectin-3 in ambulatory patients with heart failure: results from the HF-ACTION study. Circ Heart Fail 5:

19

20

21

22 Chapter 2 Galectin-3 in Cardiac Remodeling and Heart Failure Rudolf A. de Boer; Lili Yu; Dirk J. van Veldhuisen Curr Heart Fail Rep 7:1-8, 2010

23 Chapter 2 Abstract Galectin-3 is a member of the galectin family, which consists of animal lectins that bind β-galactosides. Recently, a role for galectin-3 in the pathophysiology of heart failure has been suggested. It was observed that galectin-3 is specifically upregulated in decompensated heart failure compared with compensated heart failure in animal models of heart failure. This has been associated with activation of fibroblasts and macrophages, which are a hallmark of cardiac remodeling. Therefore, galectin-3 may be a culprit biomarker in heart failure. Initial clinical observations indeed indicate that galectin-3 may be an useful biomarker for decompensated heart failure, with incremental value over well used pressure-dependent biomarkers like BNP. Future studies should be designed to deeper understanding of galectin-3 biology, in order to better address the usefulness of galectin-3 as a biomarker and to probe if anti-galectin-3 therapy may be useful in treating heart failure. Keywords: Galectin 3; Heart failure; Prognosis; Fibrosis; Macrophages; Biomarkers; Renin; Renin-Angiotensin System 22

24 Galectin-3 in Cardiac Remodeling and Heart Failure Introduction Galectins are a family of lectins that bind β-galactosides [1,2]. Galectins are expressed in vertebrates like fish, birds, amphibians, and mammals, but also in invertebrates (worms, insects) and even in lower organisms like sponge and fungus [1,2], which suggest an important role in biology. Currently, 15 members of the galectin family have been described in vertebrates. All galectins bind carbohydrate via carbohydrate-recognition domains (CRDs) with many conserved sequence elements that are typically about 130 amino acids. Some galectins contain just one CRD (galectins 1, 2, 5, 7, 10, 11, 13, 14 and 15), others (tandem-repeat like) galectins contain two homologous CRDs in a single polypeptide chain, separated by a linker of up to 70 amino acids (galectins 4, 6, 8, 9 and 12), while galectin-3 contains a non-lectin N-terminal region (about 120 amino acids) connected to a CRD and is often referred to as a chimera-like galectin. Each galectin has an individual carbohydrate-binding preference. Most galectins are bivalent or multivalent with regard to their carbohydrate-binding features: one-crd galectins may form dimers; two-crd galectins have two carbohydrate-binding sites; while galectin-3 forms pentamers upon binding to multivalent carbohydrates. So galectins are capable of forming ordered arrays made of lectin and multivalent glycoconjugates [3,4]. Galectins are primarily localized in the cytoplasm, but can also be localized in the nucleus). Galectins can be secreted, however they do so without a specific signal sequence. Galectin family members do not contain a classical signal sequence. Consistent with this feature, the proteins are localized primarily in the cytoplasm, although also in the nucleus under certain conditions. However, they can be secreted and thus belong to the group of proteins that do not contain a signal sequence but can function outside cells [5]. Galectins may bind to different cell surface antigens and receptors, in a carbohydratedependent manner. It has been suggested that galectins may not have specific individual receptors, but that each galectin can bind to a set of cell surface or extracellular matrix glycoproteins containing suitable oligosaccharides. Recent studies have demonstrated that galectins also exert intracellular functions, like signal transduction, by binding to intracellular ligands and participating in intracellular signaling pathways [6]. 2 Galectin-3: Biology and expression Biology Galectin-3 (sometimes also referred to as Mac-2, CBP-35, εbp, RL-29, HL-29, L-34, or LBP; size: 31KDa) is found in solution as a monomer with two functional domains [7-9]. Galectin-3 is unique for having an extra N-terminal domain of about amino acids long, rich in proline, glycine, tyrosine, and glutamine [10]. Galectin-3 has high affinity for lactose and N-acetyllactosamine; affinity for N-acetyllactosamine is about 5-fold higher than for lactose; ligands containing poly-n-acetyllactosamine sequences use galectin-3 as a 23

25 Chapter 2 receptor. Galectin-3 is an unique chimera-like galectin. This means galectin-3 consists of carbohydrate recognition ánd collagen-like domains which makes it capable to interact with a wide array of extracellular matrix proteins, carbohydrates (such as N-acetyllactosamine) and unglycosylated molecules, such as cell surface receptors (macrophage CD11b/CD18) and extracellular receptors (collagen IV). Galectin-3 mainly binds to glycosylated proteins of the matrix, including laminin, fibronectin and tenascin [11,12]. This has extensively been reviewed by Krzeslak et al. and Ochieng et al see the adapted Table 1 [10,13]. Table 1. Ligands for galectin-3 Extracellular matrix proteins Membrane proteins Intracellular proteins Ligand Laminin Fibronectin Tenascin M2BP Integrins: αm/β2 (CD11b/18) α1/β1 N-CAM L1 MAG LAMP-1/LAMP-2 MP20 CD98 Cytokeratins Chrp CBP70 Alix/AIP-1 Bcl-2 Gemin-4 Source /Cells EHS, macrophage, placenta Foetal Brain Brain Macrophage Adenocarcinoma Mouse brain Mouse brain Mouse brain Ubiquitious Rat lens Human T lymphoma Jurkat cells HeLa, MCF-7 Murine T3 fibroblasts HL60 Human T Lymphoma Jurkat cells Human T lymphoma Jurkat cells HeLa Others AGE Ubiquitious AGE advanced glycation end products; CBP70 glucose-binding protein; EHS Engelbroth-Holm-Swarm tumor laminin; LAMP lysosomal-associated membrane protein 1; M2BP Mac-2 binding protein; MAG myelinassociated glycoprotein; MP20 membrane protein 20; NCAM neural cell adhesion molecule (adapted from Krzeslak [10]). 24

26 Galectin-3 in Cardiac Remodeling and Heart Failure Galectin-3 is localized in the cytoplasma and the nucleus. Loss of nuclear galectin-3 is believed to affect the malignant phenotype of cancers. Nuclear expression of galectin-3 is associated with proliferative effects and it has been described that galectin-3 translocates from the cytosol into the nucleus via a passive and an active pathway [14]. Galectin-3 contains several phosphorylation sites and other determinants important for the secretion of galectin-3, which goes via a novel, non classical pathway [15]. Through its interaction with extracellular matrix proteins galectin-3 is able to cross the plasma membrane despite its lack of appropriate signal peptides; secretion of galectin-3 is critically regulated at the plasma membrane [16]. Expression of Galectin-3 The different members of the galectin family exhibit a specific pattern of expression in various cells and tissues. Expression of galectin-3 has been detected in macrophages, eosinophils, neutrophils, and mast cells [17,18]. In tissues, galectin-3 is most abundantly expressed in lung, spleen, stomach, colon, adrenal gland, uterus and ovary [19]. Galectin-3 is also expressed, albeit at a much lower level, in kidney, heart, cerebrum, pancreas, and liver [19]. However, under pathophysiological conditions, the level of expression of gelactin-3 may change substantially so that a low constitutive expression level of galectin-3 does not preclude a prominent function, e.g. in liver and heart. Galectin-3 contains several phosphorylation sites and other determinants important for the secretion of galectin-3, which goes via a novel, non classical pathway [15]. Secretion of galectin-3 is critically regulated at the plasma membrane [16]. 2 Galectin-3 in Cardiac Remodeling and Heart Failure Experimental observations The observation that galectin-3 is increased in decompensated heart failure has been published in a seminal paper by Sharma [20]. The authors studied homozygous REN2- rats that overexpress the murine 2-d renin gene, resulting in severe hypertension with end-organ damage [21-22]. It was observed that while some rats go into overt failure after about 15 weeks, with signs of heart failure like dyspnea, lethargy and severely compromised hemodynamics, some rats remain compensated. These two groups were compared using a cdna array with whole RNA from the rat hearts. It was observed that a set of 48 genes were differentially regulated [20,23]. Interestingly, most of the differentially regulated genes typically encoded matricellular proteins, like collagens, osteoactivin and fibronectin, but not loading-dependent factors like natriuretic peptides [23]. Galectin-3 was the strongest regulated gene, being upexpressed in decompensated hearts more than 5-fold compared to compensated hearts. 25

27 Chapter 2 To dissect cause from consequence, Sharma and colleagues showed that infusion with galectin-3 in the pericardial sac of normal rats led to the development of cardiac remodeling with dysfunction and increased expression of collagens. Given the upregulation of galectin-3 well before the transition to overt heart failure, the authors concluded that galectin-3 may be factor that should be considered as a novel target for intervention in heart failure. A recent paper by Thandavarayan [24] described a model with cardiospecific expression of a dominant-negative form of protein, which regulates apoptosis and several signaling pathways that leads to LV dysfunction. Besides typical changes for LV remodeling, like hypertrophy, fibrosis, and apoptosis, the authors also describe an upregulation of galectin- 3 in the LV [24]. So, upregulation of galectin-3 may be a general phenomenon in LV dysfunction and not be confined to models with increased AngII signaling. A potential role in mediating the effects of galectin-3 has been suggested for N-acetyl- Ser-Asp-Lys-Pro (ac-sdkp), a tetrapeptide degraded by angiotensin-converting enzyme (ACE) by the group of Carretero [25,26]. First, they showed that differentiation of murine bone marrow cells to macrophages was inhibited by ac-sdkp. Second, in mice treated with AngII, ac-sdkp reduced fibrosis and expression of galectin-3 in left ventricular tissue [25]. In a second study, LV remodeling was induced by infusion of galectin-3 in the pericardial sac [26], with or without the co-administration of ac-sdkp. Like in the initial report by Sharma [20], galectin-3 enhanced macrophage and mast cell infiltration which is associated with interstitial and perivascular fibrosis and LV dysfunction. Ac-SDKP prevented these events, in whole or in part, and these effects were shown to be mediated by TGF-β/Smad3 pathway. Fibrosis Galectin-3 seems to be particularly involved in fibrosis. Fibrosis and scar formation are pivotal processes in maladaptive cardiac remodeling. Fibroblasts, myofibroblasts, and macrophages have been identified as important cells in the initiation and progression of tissue scarring [27-29]. Various fibrotic conditions are associated with up-regulation of galectin-3: liver cirrhosis [30,31], idiopathic lung fibrosis [32], and chronic pancreatitis [33]. In animal models upregulation of galectin-3 has been described for hepatic [31], renal [34-36], and cardiac [20,25,26] fibrosis. More detailed study into the role of galectin-3 in cardiac remodeling revealed that galectin-3 was localized at the very sites of fibrosis, colocalizing with fibroblasts and macrophages, but not with cardiomyocytes. Galectin-3 binding sites were visualized by biotynalyted antibodies and localized predominantly to fibrotic areas [20], in line with the evidence that galectin-3 binds extracellular proteins. Furthermore, recombinant galectin-3 causes proliferation and collagen production of cultures cardiac fibroblasts in vitro. Other published data corroborate these observations [25,26]. Galectin-3 has also been identified as a potentially important mediator of removal of advanced glycosylation end-products (AGEs) [37]. AGEs are molecules formed during a non- 26

28 Galectin-3 in Cardiac Remodeling and Heart Failure enzymatic reaction between proteins and sugar residues, and accumulate in the human body with age, but also with enhanced states of oxidative stress, an increased intake of AGEs, and renal and heart failure, and are thought to contribute to myocardial stiffening [38]. An AGEspecific cellular receptor complex (AGE-R) mediating AGE removal has been described. Galectin-3 interacts with AGE-R components thus contributing to the elimination of these pathogenic substances. Galectin-3 disruption is associated with increased susceptibility to AGE-induced renal disease, which indicates that galectin-3 is operating in vivo as an AGE receptor to afford protection toward AGE-dependent tissue injury [39,40]. Whether galectin-3 acts on AGE biology in the heart remains unknown. Other experimental lines of evidence provide further compelling evidence that galectin-3 is a key player in fibrosis. Especially the generation of galectin-3 deficient mice has been instrumental in the study of galectin-3 in fibrosis. The group from Sehti has published papers on the role of galectin-3 in hepatic and renal fibrosis [31,35]. In the liver, TGF-β via galectin- 3 activates myofibroblasts. In the kidney, galectin-3 expression and secretion by macrophages is a major mechanism in renal fibrosis. 2 Inflammation Besides fibrosis, galectin-3 plays an important role in the inflammatory response, which is an important player in the process of cardiac remodeling [41]. In renal models of inflammation, galectin-3 has been convincingly linked to fibrosis and damage [34,36]. Employing a model of murine renal fibrosis, Henderson and colleagues established that depletion of macrophages significantly reduced myofibroblast activation and decreased fibrosis [35]. In the heart, no direct evidence exists has been generated for inflammatory effects via galectin-3, although in Sharma s paper besides galectin-3 other genes encoding for immune factors were differentially regulated [20]. Mice which constitutively express IFNgamma in their livers develop myocarditis with macrophages expressing high levels of galectin-3 [42]. Clinical Studies The first report in human subjects was the paper by Sharma et al. [20]. They studied ventricular biopsies from patients with aortic stenosis with preserved or depressed ejection fraction and showed that gealctin-3 was upregulated in the biopsies from patients with depressed ejection fraction. Van Kimmenade et al. [43] published the first clinical study that evaluated the potential role of galectin-3 as a plasma biomarker in heart failure. In this study, 599 acutely dyspneic subjects were evaluated with the goal to establish the usefulness of NTproBNP, galectin-3 and apelin in diagnosing heart failure and predicting outcome. A blood sample was collected at baseline, and NT-proBNP, galectin-3 and apelin were measured later. A total of 209 patients were diagnosed with heart failure. NT-proBNP came up as the most 27

29 Chapter 2 powerful predictor for diagnosing heart failure. ROC analysis examining the value of NTproBNP for the diagnosis acute heart failure showed an area under the curve (AUC) for NTproBNP of 0.94 (P<0.0001), while the AUC for galectin-3 for the diagnosis acute heart failure was 0.72 (P<0.0001) the difference between NT-proBNP and galectin-3 being highly significant (P<0.0001). The optimal cut-off of galectin-3 in this study was 6.88 ng/ml, which resulted in a reasonable sensitivity of 80% but a poor specificity of 52% [43]. For predicting short term prognosis (60 days, primary endpoint rehospitalization due to heart failure (N=60) or all cause mortality (N=17), gealectin-3 was the most powerful predictor: an AUC for galectin-3 of 0.74 (P=0.0001) and an AUC for NT-proBNP of 0.67 (P=0.009); the difference being borderline significant (P=0.05). In multivariate analysis, galectin-3 was the strongest predictor for death and the combination of death and rehospitalizations for heart failure. Remarkably, well known predictors for outcome like NT-proBNP and renal function were not predictive in this study. Nevertheless, this study provides strong support for the exploration of galectin-3 as a biomarker that may predict prognosis, while its usefulness in detecting heart failure or adding incremental value (over currently used clinical correlates and NT-proBNP) in the diagnostic work-up of heart failure remains unclear. In a larger study in patients with chronic heart failure, Lok et al. [44] showed that galectin-3 predicts long-term outcome (mean follow-up: 3.4 years); HR 1.95, CI , P=0.004). In this study, not many other biomarkers of heart failure were measured, so that is impossible to value the precise role of galectin-3 in this cohort. An interesting mechanistic study by Milting and colleagues [45] describes the kinetics of galectin-3 in 55 patients with end stage heart failure with the need for mechanical circulatory support or MCS. In this small study several biomarkers, especially related to myocardial fibrosis and remodeling, were determined. First, the fibrosis-related biomarkers including gelaectin-3 were increased compared with controls. Second, the authors reported that none of the fibrosis-related biomarkers like TIMP1, tenascin, osteopontin or galectin-3 were reduced by MCS; only the loading-related biomarker BNP was reduced by MCS. Third, patients who did not survive while on MCS compared with patients who lived until transplantation had higher baseline galectin-3 levels. A recent study by Lin et al. [46] describes the relation between serum galectin-3 and markers of extracellular matrix turnover. They studied 106 patients with chronic heart failure (NYHA class II-III, mean LVEF 35±9%). Serum aminoterminal propeptide of procollagen (type I, PINP) and type III (PIIINP), matrix metalloproteinase-2(mmp-2) and tissue inhibitor of metalloproteinase-1 (TIMP-1) were analyzed, along with galectin-3. Galectin-3 was correlated with PIIINP, TIMP-1, MMP-2, but not with LVEF, age and sex. After correction, the correlation between galectin-3 and PIIINP and MMP-2 remained statistically significant. The authors conclude these findings suggest a relationship between gelactin-3 and extracellular matrix turnover. Taken together, from available clinical data, plasma and/or serum galectin-3 is increased in acute and chronic heart failure. It seems that galectin-3 may be of particular value to predict 28

30 Galectin-3 in Cardiac Remodeling and Heart Failure prognosis, as for clinical diagnosing and/or decision making it seems less powerful, although we do not have sufficient data available. Conclusions Galectin-3 is interesting and complex protein with many effects in many organs. In the failing or stressed heart, it has been shown that activated macrophages secrete galectin-3. The increased expression levels of galectin-3 are associated with the tendency to develop decompensated heart failure, and in clinical cohorts, increased plasma galectin-3 levels are linked with worse prognosis. Therefore, galectin-3 may be advocated as a novel biomarker, but it may also be in the pathophysiological circle of heart failure ( culprit biomarker ), and therefore it may also be target for intervention [47]. The suggested pathways of galectin-3 are displayed in figure 1 [47]. 2 Figure 1. Galectin-3 pathways. The network represents molecular relationships between different gene products. Node shapes indicate the functional class of the gene product while node colours indicate a role in general fibrosis (orange) or cardiac fibrosis (green). Edge colurs indicate up-regulation or activation (red), down-regulation or inhibition (green) or involvement without clear directionality (yellow). All relationships are supported by references from the ingenuity Pathway Knowledge Base or key reference included in the review by de Boter et al. [47]. Figure and legend reprinted with permission of the publisher (Oxford Journals; Eur J Heart Fail.2009, 11: ). There are several fields of uncertainty. First, we do not know how galectin-3 is regulated at a transcriptional and translational level in the heart. Mechanistic studies have provided evidence that cardiac fibroblasts and macrophages are the main sources for galectin-3, and that the TGF-β/smad pathway is involved. Inflammatory signals also contribute in the regulation of galectin-3. However, what precise signals govern the production and secretion of 29

31 Chapter 2 galectin-3 remains largely enigmatic. Second, although several lines of evidence strongly suggest a contributory role for galectin-3 in the pathophysiology of heart failure, we lack proof-of-principle experiments, e.g. in galectin-3 deficient mice or in pharmacological studies, to show that galectin-3 is unequivocally contributing to the onset and progression of cardiac remodeling. Finally, we have no data if therapy, or what therapy, may affect galectin-3 expression and signaling. Taken together, from available clinical data, plasma and/or serum galectin-3 is increased in acute and chronic heart failure. It seems that galectin-3 may be of particular value to predict prognosis in HF; the cause for clinical diagnosing and/or decision making is less convincing; to date we do not have data available to support the use of galectin-3 for this purpose. Parallel to the experiments described by Sharma [20], we have observed that galectin-3 is upregulated in compensated and decompensated hypertrophy when compared with healthy control rats, however the difference is of lesser extent than when comparing the decompensated versus the compensated rats (de Boer, unpublished data). This suggests that galectin-3 may be of lesser importance during the early stages of the disease. Furthermore, no data on serial measurements of galectin-3 have been published. So, we lack data on half-life, kinetics, clearance and other parameters of galectin-3 biology. The study from Milting [45] suggests that acute unloading by MCS only reduces typical loading biomarkers, i.e. BNP (or NTproBNP), not so much biomarkers associated with turnover of the extracellular matrix, including galectin-3. Most treatment modalities in heart failure however, are not acute interventions but chronic pharmacological neurohormonal inhibition [48]. However, if such long term treatment aimed at reduction of matrix apposition and fibrosis, like ACE-inhibition and aldosteron receptor blockers will result in lowering of galectin-3 levels is currently unknown and warrants further study. If it is indeed proven that standard treatment of heart failure is associated with lowering of galectin-3 expression and levels, one could argue that galectin-3 itself could also be a target for therapy. Specific agents targeted against galectins have been tested in small trials with cancer patients. These agents have not been evaluated in experimental or clinical heart failure, but will likely be tested momentarily for their value in this devastating disease. Acknowledgement This work was supported by the Netherlands Heart Foundation (grant 2007T046 to R.A.d.B.) and the Innovational Research Incentives Scheme program of the Netherlands Organization for Scientific Research (NWO VENI, grant to R.A.d.B.). 30

32 Galectin-3 in Cardiac Remodeling and Heart Failure Reference 1. Barondes SH, Cooper DNW, Gitt MA, Leffler H. Galectins. Structure and function of a large family of animal lectins. J Biol Chem 1994, 269: Cooper DN. Galectinomics: finding themes in complexity. Biochim Biophys Acta 2002, 1572: Wang JL, Laing JG, Anderson RL. Lectins in the cell nucleus. Glycobiology 1991, 3: Yang RY, Rabinovich GA, Liu FT. Galectins: structure, function and therapeutic potential. Expert Rev Mol Med 2008, 13:e17 e Elola MT, Wolfenstein-Todel C, Troncoso MF, Vasta GR, Rabinovich GA. Galectins: matricellular glycan-binding proteins linking cell adhesion, migration, and survival. Cell Mol Life Sci 2007, 64: Liu, F.T., Patterson, R.J. and Wang, J.L. Intracellular functions of galectins. Biochim Biophys Acta 2002, 1572: Wang JL, Laing JG, Anderson RL. Lectins in the cell nucleus. Glycobiology 1991, 3: Hughes RC. Mac-2: a versatile galactose-binding protein of mammalian tissues. Glycobiology 1994, 4: Birdsall, B, Feeney J, Burdett IDJ, et al. NMR solution studies of hamster galectin-3 and electron microscopic visualization of surface-adsorbed complexes: evidence for interactions between the N- and C-terminal domains. Biochemistry 2001, 40: Krześlak A, Lipińska A. Galectin-3 as a multifunctional protein. Cell Mol Biol Lett 2004, 9: Rosenberg I, Cherayil BJ, Isselbacher KJ, Pillai S. Mac-2-binding glycoproteins. Putative ligands for a cytosolic β- galactoside lectin. J Biol Chem 1991, 266: Sato S, Hughes RC. Binding specificity of a baby hamster kidney lectin for H type I and II chains, polylactosamine glycans, and appropriately glycosylated forms of laminin and fibronectin. J Biol Chem 1992, 267: Ochieng J, Furtak V, Lukyanov P. Extracellular functions of galectin-3. Glycoconj J 2004, 19: Nakahara S, Oka N, Wang Y, et al.. Characterization of the nuclear import pathways of galectin-3. Cancer Res 2006, 66: Menon RP, Hughes RC. Determinants in the N-terminal domains of galectin-3 for secretion by a novel pathway circumventing the endoplasmic reticulum-golgi complex. Eur J Biochem 1999, 264: Mehul B, Hughes RC. Plasma membrane targetting, vesicular budding and release of galectin 3 from the cytoplasm of mammalian cells during secretion. J Cell Sci 1997, 110: Hughes RC. The galectin family of mammalian carbohydrate-binding molecules. Biochem Soc Transact 1997, 25: Hughes RC. Secretion of the galectin family of mammalian carbohydrate-binding family proteins. Biochem Biophys Acta 1999, 1473:

33 Chapter Kim H, Lee J, Hyun JW, et al. Expression and immunohistochemical localization of galectin-3 in various mouse tissues. Cell Biol Int 2007, 31: Sharma UC, Pokharel S, van Brakel TJ, et al. Galectin-3 marks activated macrophages in failure-prone hypertrophied hearts and contributes to cardiac dysfunction. Circulation 2004, 110: Lee MA, Böhm M, Paul M, et al. Physiological characterization of the hypertensive transgenic rat TGR(mREN2)27. Am J Physiol 1996, 270:E de Boer RA, Pokharel S, Flesch M, et al. Extracellular signal regulated kinase and SMAD signaling both mediate the angiotensin II driven progression towards overt heart failure in homozygous TGR(mRen2)27. J Mol Med 2004, 82: Schroen B, Heymans S, Sharma U, et al. Thrombospondin-2 is essential for myocardial matrix integrity: increased expression identifies failureprone cardiac hypertrophy. Circ Res 2004, 95: Thandavarayan RA, Watanabe K, Ma M, et al protein regulates Ask1 signaling and protects against diabetic cardiomyopathy. Biochem Pharmacol 2008, 75: Sharma U, Rhaleb NE, Pokharel S, et al. Novel anti-inflammatory mechanisms of N-Acetyl-Ser-Asp-Lys-Pro in hypertension-induced target organ damage. Am J Physiol 2008, 294:H1226 H Liu YH, D'Ambrosio M, Liao TD, et al. N-acetyl-seryl-aspartyl-lysyl-proline prevents cardiac remodeling and dysfunction induced by galectin-3, a mammalian adhesion / growth-regulatory lectin. Am J Physiol Heart Circ Physiol 2009, 296:H Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 2000, 275: Brown RD, Ambler SK, Mitchell MD, Long CS. The cardiac fibroblast: therapeutic target in myocardial remodeling and failure. Annu Rev Pharmacol Toxicol 2005;45: de Cavanagh EM, Ferder M, Inserra F, Ferder L. Angiotensin II, mitochondria, cytoskeletal, and extracellular matrix connections: an integrating viewpoint. Am J Physiol Heart Circ Physiol 2009, 296(3):H Hsu DK, Dowling CA, Jeng KC, et al.. Galectin-3 expression is induced in cirrhotic liver and hepatocellular carcinoma. Int J Cancer 1999, 81: Henderson NC, Mackinnon AC, Farnworth SL, et al. Galectin-3 regulates myofibroblast activation and hepatic fibrosis. Proc Natl Acad Sci USA 2006, 103: Nishi Y, Sano H, Kawashima T, et al. Role of galectin-3 in human pulmonary fibrosis. Allergol Int 2007, 56: Wang L, Friess H, Zhu Z, et al. Galectin-1 and galectin-3 in chronic pancreatitis. Lab Invest 2000, 80:

34 Galectin-3 in Cardiac Remodeling and Heart Failure 34. Sasaki S, Bao Q, Hughes RC. Galectin-3 modulates rat mesangial cell proliferation and matrix synthesis during experimental glomerulonephritis induced by anti-thy1.1 antibodies. J Pathol 1999, 187: Henderson NC, Mackinnon AC, Farnworth SL, et al. Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol 2008, 172: Eis V, Luckow B, Vielhauer V, et al. Chemokine receptor CCR1 but not CCR5 mediates leukocyte recruitment and subsequent renal fibrosis after unilateral ureteral obstruction. J Am Soc Nephrol 2004, 15: Vlassara H, Li YM, Imani F, et al. Identification of galectin-3 as a high-affinity binding protein for advanced glycation end products (AGE): a new member of the AGE-receptor complex. Mol Med 1995, 1: Hartog JW, Voors AA, Bakker SJ, Smit AJ, van Veldhuisen DJ. Advanced glycation end-products (AGEs) and heart failure: pathophysiology and clinical implications. Eur J Heart Fail 2007, 9: Iacobini C, Oddi G, Menini S, et al. Development of age-dependent glomerular lesions in galectin-3/age-receptor-3 knockout mice. Am J Physiol 2005, 289:F Iacobini C, Menini S, Oddi G, et al. Galectin-3/AGE-receptor 3 knockout mice show accelerated AGE-induced glomerular injury: evidence for a protective role of galectin-3 as an AGE receptor. FASEB J 2004, 18: Frangogiannis NG. The immune system and cardiac repair. Pharmacol Res 2008, 58: Reifenberg K, Lehr HA, Torzewski M, et al. Interferon-gamma induces chronic active myocarditis and cardiomyopathy in transgenic mice. Am J Pathol 2007, 171: van Kimmenade RR, Januzzi JL Jr, Ellinor PT, et al. Utility of aminoterminal pro-brain natriuretic peptide, galectin-3, and apelin for the evaluation of patients with acute heart failure. J Am Coll Cardiol 2006, 48: Lok D, van der Meer P, de La Porte PB, et al. Galectin-3, a novel marker of macrophage activity, predicts outcome in patients with stable chronic heart failure. J Am Coll Cardiol 2007, 49(Suppl. A): 98A[Abstract]. 45. Milting H, Ellinghaus P, Seewald M, et al. Plasma biomarkers of myocardial fibrosis and remodeling in terminal heart failure patients supported by mechanical circulatory support devices. J Heart Lung Transplant 2008, 27: Lin YH, Lin LY, Wu YW, et al. The relationship between serum galectin-3 and serum markers of cardiac extracellular matrix turnover in heart failure patients. Clin Chim Acta 2009, 409: de Boer RA, Voors AA, Muntendam P, van Gilst WH, van Veldhuisen DJ. Galectin-3: a novel mediator of heart failure development and progression. Eur J Heart Fail 2009, 11: Dickstein K, Cohen-Solal A, Filippatos G, et al. ESC guidelines for the diagnosis treatment of acute, chronic heart failure The task force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur J Heart Fail 2008, 10:

35

36

37

38 Chapter 3 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure Lili Yu ; Rudolf A. de Boer Book: CARDIAC REMODELING: MOLECULAR MECHANISMS Edited by Bodh I. Jugdutt and Naranjan S. Dhalla, 2012

39 Chapter 3 Abstract Myocardial injuries stemming from pressure overload or myocardial infarction lead to cardiac remodeling and represent major health problems world-wide. An ever accumulating body of experimental and clinical research appoints galectin-3, a β-galactoside-binding lectin, as a key player in this maladaptive response to myocardial injury. Herein, a specific role for galectin-3 in inflammation and fibrogenesis has been elucidated in experimental and clinical studies. Galectin-3 was first associated with pathological conditions leading to cardiac remodeling, such as inflammation and fibrosis. Then, as the carbohydrate recognition domain of galectin-3 reacts with glycosylated proteins such as laminin, fibronectin and tenascin, a multifunctional role of galectin-3 in the extracellular matrix was postulated. Notably, experimental animal studies clearly showed that galectin-3 is a mediator of crucial steps in fibrogenesis, and further induces cardiac inflammation, hypertrophy and dysfunction. Possible mechanisms pertaining to galectin-3 inflammatory and fibrotic properties have been suggested to involve macrophage activation, galectin-3-induced chemotaxis and activation of the TGFβ-Smad3 signaling pathways. Additionally, the link between plasma galectin-3 and fibrosis was also established in clinical biomarker studies. Galectin-3 and its pathways may be explored further in order to develop more efficient strategies to target cardiac remodeling in heart failure leading to fibrosis. Key words: Galectin-3 heart failure remodeling biomarker 38

40 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure STRUCTURE, EXPRESSION AND FUNCTION OF GALECTIN-3 Galectin-3 is a kda chimaera-type galectin. Galectins form a large family of galactosidase binding lectins, but galectin-3 is the only member of the galectin family with an extended N-terminal domain ( amino acids). This N-terminal domain is composed of tandem repeat sequences comprising nine amino acid residues and is connected to a C- terminal carbohydrate recognition domain (CRD) of about 130 amino acids [1, 2]. The CRD interacts with various glycosylated proteins, modulates cell-cell adhesion and cell signaling in the extracellular compartment [1]. Further, the galectin-3 CRD and collagen-like domains communicate with a variety of extracellular matrix proteins (ECM), carbohydrates (e.g., N- acetyllactosamine), and also unglycosylated proteins such as cell surface receptors (macrophage antigens CD11b/CD18) and extracellular receptors (collagen IV) [3-6], and they furthermore modulate cell-cell adhesion and cell signaling in the extracellular compartment. Galectins are expressed in various cells and tissues and are important for diverse physiological and pathological processes, such as immune and inflammatory responses, tumor development and progression, neural degeneration, atherosclerosis, diabetes, as well as wound repair. Galectin-3 has been detected in many proliferative cells including tumor cells, eosinophils, neutrophils, and activated macrophages [7, 8]. Notably, many of these cells types are operative in the inflammatory response and the formation of fibrosis (fibrogenesis). Galectin-3 is predominantly located in the cytoplasm. While intracellular galectin-3 plays a pivotal role in diverse cell growth, anti-apoptosis signaling, and mrna splicing, extracellular (or cell surface bound) galectin-3, on the other hand, participates in cell-cell and cell-matrix adhesion, by binding to glycosylated ECM components, including laminin, fibronectin, tenascin, and Mac-2 binding protein [3, 4, 9-12]. Thus, extracellular galectin-3 appears in tight communication with factors involved in fibrogenesis. Differential expression of galectin-3 has been reported for different murine organs. Herein, lower expression was found in cerebrum, heart, and pancreas, while moderate expression was found in liver, ileum, kidney and adrenal gland and high expression in lung, spleen, stomach, colon, uterus, and ovary [13]. Figure 1 displays an overview of galectin-3 distribution in various healthy tissues of different species our data are in concert with results published by Kim and colleagues [13]. Notably, a growing body of evidence reveals that high expression levels of galectin-3 are closely associated with pathological conditions, specifically conditions pertaining to inflammation and fibrosis which are also key conditions in cardiac remodeling. 3 Localization of galectin-3 in damaged tissue Inflammation and fibrosis are intertwined pathological states and galectin-3 has consistently been observed to be involved in these damage states. Galectin-3 is not only secreted by activated macrophages but in situ hybridization and immunohistochemistry analysis showed that galectin-3 is highly distributed in the fibrotic 39

41 Chapter 3 Figure 1: Immunohistochemical localization, protein and mrna expression of galection-3 in various tissues of different species. Galectin-3 immunoreactivity has been observed in heart (A), Lung (B), Liver (C), Spleen (D), and Kidney (E). F-I: Galectin-3 protein expression (assessed by Western blot) and mrna expression (assessed by qpcr) has been predominantly observed in (mouse, rat, human) lung, testis/ovary, spleen, prostate, adipose tissue en skin. Picture A E reprinted from reference [13] with permission from the publisher; Figure F I: unpublished data (Yu et al.). Scale bars, 50 µm ( A and B); 60 µm (C); 90mm (D); 400 µm(e). area of myocardium and co-localizes to macrophages [14]. Moreover, an ever growing body of experimental evidence found that galectin-3 manifests in various damaged tissue, in particular in tissues with increased collagen deposition and localization in the fibrotic area [14-23] Further, galectin-3 was expressed in proliferating fibroblast and was also found in nucleus when cells were exposed to apoptotic stimuli in vitro [24]. Moreover, galectin-3 expression has been detected in isolated cardiac fibroblast to localize galectin-3 binding sites. Interestingly, contrary to cardiac fibroblast, galectin-3 binding sites were absent from cardiomyocytes [7, 8, 14, 25-27]. GALECTIN-3 IN EXPERIMENTAL FIBROSIS Early experimental support in liver, kidney and lung fibrosis 40

42 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure A prominent role for galectin-3 in fibrogenesis has been elucidated in recent years. Herein, work has specifically described fibrotic states in the liver, kidneys, pancreas and lungs. For example, galectin-3 showed to be temporarily and spatially associated with fibrosis with a minimal expression in healthy rat liver, while highest expression was found at peak fibrosis and virtually no galectin-3 expression was present after recovery from fibrosis [19]. Then, galectin-3 deficiency in bile duct ligated (BDL) rats was recently shown to significantly diminish BDL induced hepatic TGF-β1 and procollagen expression and associated hepatic fibrosis [28]. Additionally, in a murine model of cirrhosis, bone marrow cell transplantation significantly decreased liver fibrosis which was associated with decreased hepatic galectin-3 expression [29]. A recent report by Mackinnon et al [23] extended the involvement of galectin-3 to fibrosis in the lung. Authors found a significantly reduced TGF-β1 and bleomycin-induced lung fibrosis in galectin-3 deficient mice. Likewise, galectin-3 expression and secretion by macrophages has been identified as a major contributor to renal fibrosis. Mice with galectin-3 deficiency did not show macrophage recruitment upon interferon-gamma/lps stimulation [19]. Henderson et al. also observed that galectin-3 mediated TGF-β induced myofibroblast activation, a crucial step in the fibrogenesis cascade [17]. Lastly, the therapeutic potential of targeting galectin-3 to relieve fibrotic burden has recently been investigated by Kolatsi-Joannou and colleagues [30] who demonstrated that folic acid induced kidney fibrosis and associated galectin-3 expression was significantly reduced by 1% treatment with modified citrus pectin (MCP). MCP is a pectin derivative binding to the galectin-3 CRD thereby inhibiting galectin-3 aforementioned effects. Altogether, a growing body of literature consistently supports a role of galectin-3 in fibrogenesis. This appears a generalized effect, not confined to one organ. Fibrosis is also central to the maladaptive response to myocardial injury, such as pressure overload and myocardial infarction leading to cardiac remodeling [14, 21, 22] Supporting the notion that galectin-3 constitutes a global player in fibrogenesis, it is not surprising that galectin-3 has also been shown to be involved in myocardial fibrogenesis. 3 Galectin-3 in cardiac remodeling pathway recent experimental support The first evidence showing an involvement of galectin-3 in heart failure stems from a landmark study of Sharma and colleagues [14]. In a comprehensive microarray study galectin- 3 was identified as the most robustly overexpressed gene in heart-failure prone hypertrophied hearts compared to functionally compensated hearts in homozygous transgenic TGFmRen2-27(Ren-2) rats. Ren-2 rats overexpress the mouse Ren-2d renin gene and spontaneously develop heart failure after weeks. Further, continuous low-dose infusion of recombinant galectin-3 into the pericardial sac caused left ventricular dysfunction in healthy Sprague- Dawley rats, associated with collagen deposition and other signs of cardiac remodeling [14]. These initial observations warranted galectin-3 to be considered as a new target for intervention in heart failure. 41

43 Chapter 3 Then, more recently, Sharma et al [21] found a significant and high expression of galectin-3 in cardiac tissue of Ang-II induced hypertension in mice. In this model, galectin-3 was released by infiltrating macrophages in the myocardium and led to interstitial collagen deposition. Treatment with N-Acetyl-Ser-Asp-Lys-Pro (Ac-SDKP), an endogenous tetrapeptide specifically degraded by angiotensin converting enzyme (ACE), reduced macrophage activation and galectin-3 expression herein, and prevented galectin-3-induced cardiac inflammation, fibrosis and remodeling. Additionally, Liu et al showed that in galectin-3-induced cardiac remodeling, galectin-3 increased the number of macrophages, mast cell infiltration and activated the TGF β/smad3 pathway. Ac-SDKP partially or near completely prevented galectin-3 induced cardiac inflammation, fibrosis, hypertrophy and dysfunction, possibly by inhibition of the TGFβ/Smad3 signaling pathway [22]. Recent studies by Thandavaryan et al. [31], Kamal et al. [32] and by Psarras et al. [33] further reveal a specific association of galectin-3 with left ventricular dysfunction and fibrosis. Specifically, Thandavaryan et al [31] found a significant increase in myocardial hypertrophy and fibrosis, as well as apoptosis, all associated with up-regulated galectin-3 in η protein mutant (DN η) mice after induction of diabetes η protein has a regulatory role in apoptosis, adhesion, cellular proliferation, differentiation, survival and signal transduction pathways [34]. As the authors stated, up-regulated galectin-3 appears to be a general phenomenon in LV dysfunction [31]. Moreover, Kamal et al. [32] demonstrated that cardiac hypertrophy, progressive fibrotic cardiac remodeling with increased collagen deposition were accompanied with significantly increased galectin-3 expression in cardiomyopathic hearts in a rat model of myosin-induced experimental autoimmune myocarditis (EAM). Notably, herein galectin-3 over-expression was dramatically reduced by treatment with T-3999, a novel phenylpyridazinone, indicating an inhibitory function of T on galectin-3. Then, Psarras et al. [33] found that desmin deficient (des-/-) mice exhibit marked myocardial degeneration and fibrosis, which were associated with osteopontin (OPN) and galectin-3 overexpression (226x for OPN and 26x for galectin-3). OPN, like galectin-3, has chemotactic properties and is thus recruited to inflammatory sites [35, 36]. Psarras et al. [33] further compared des-/- OPN -/- mice with des-/- OPN +/+ mice and found that des-/- OPN -/- mice not only displayed remarkable improvements in ventricular function (53%) but also in myocardial fibrosis (44%) while also significantly reducing galectin-3 gene expression (by 80%) compared to des-/- OPN +/+ mice, indicating that the observed diminished inflammatory and fibrotic response in OPN deficient des-/- mice could be partly explained by the significant reduced myocardial galectin-3 level [33]. So, accumulating experimental evidence implicates a role of galectin-3 in the development of organ fibrosis. Whether galectin-3 is a potential therapeutic target in left ventricular (LV) remodeling and heart failure is unknown. We have conducted experimental studies and our results suggest galectin-3 may a target for therapy. Genetic disruption and 42

44 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure pharmacological inhibition of galectin-3 attenuated adverse LV remodeling, fibrosis and subsequent HF development. We perturbed mice with angiotensin II (AngII) and transverse aortic constriction (TAC) causing LV hypertrophy, decreased LV contractility and increased LV end-diastolic pressure, associated with increased fibrosis in wild type (WT) mice. However, galectin-3 knock out (Gal3-KO) mice did not develop LV dysfunction and fibrosis. Additionally, in homozygous TGR(mREN)27 rats, pharmacological inhibition of galectin-3 with an oligosaccharide almost completely prevented LV dysfunction and fibrosis [54, 55]. This indeed suggests that drugs binding to galectin-3 may be potential therapeutic candidates for the prevention of heart failure with extensive fibrosis. It remains unclear what mechanisms underpin these effects. ROLE OF GALECTIN-3 IN MODULATION OF FIBROSIS Potential mechanism of galectin-3 in extracellular matrix Fibroblasts, myofibroblasts and macrophages have been identified as important cells in the initiation and progression of fibrogenesis, scar formation, and tissue remodeling [37-39]. Extracellular matrix remodeling (ECM) is a crucial aspect in fibrogenesis and galectin-3 seems to play a multifunctional role in the ECM environment, as its CRD reacts with glycosylated proteins in the ECM, such as laminin, fibronectin, tenascin [3, 4], as well as membrane proteins, such as αm/β2 (CD11b/18) [40]. 3 Potential mechanisms of galectin-3 in myofibroblast differentiation Fibroblast to myofibroblast differentiation and activation by inflammatory cytokines, like TGF-β, preceded by influx of cells such as macrophages are some of the initial steps in the process of fibrogenesis. A large body of research supports a role for galectin-3 in this process. Herein, it appears that macrophages and TGF-β induce myofibroblast activation via galectin- 3, but that macrophage recruitment and TGF-β expression is independent of galectin-3. First, galectin-3 was visualized in the fibrotic area co-localizing with fibroblasts and macrophages [14]. Second, it was shown that infusion of recombinant galectin-3 into pericardial sac leads to inflammatory cell infiltration, cardiac fibroblast proliferation, collagen synthesis and deposition, essentially contributing to interstitial and perivascular fibrosis [14, 21, 22]. Then, Dvorankova et al. demonstrated myofibroblast activation in vitro upon treatment with a moderately high dose recombinant galectin-3 [41]. Further, galectin-3 deficiency significantly reduced myofibroblast activation in carbon tetrachloride (CCL4) induced hepatic fibrosis and renal fibrosis (in a model of unilateral ureter obstruction, UUO) [17, 19]. Then, macrophages regulate fibroblast and myofibroblast activation in ECM and macrophage derived galectin-3 presents in various fibrotic pathologies. Injured tissue displays a marked increase in galectin-3 expression by activated macrophages and also an increased TGF-β expression, all these factors promote fibroblast proliferation and myofibroblast activation [17]. Macrophage depletion, then, significantly inhibits myofibroblast activation 43

45 Chapter 3 and decreases fibrosis [19]. For example, Henderson and colleagues showed that galectin-3 deficient macrophage recruitment could not drive myofibroblast accumulation and activation. Utilizing a cross-over experiment with wild-type or galectin-3 deficient macrophage supernatant and galectin-3 deficient renal fibroblast, these authors further observed that proliferation of galectin-3 deficient renal fibroblasts were activated by wild-type macrophages and attenuated by a galectin-3 inhibitor bis-(3-deoxy-3-{3-methoxybenzamido}-β-dgalactopyranosyl-sulfane), while galectin-3 deficient macrophages did not induce proliferation in galectin-3 deficient renal fibroblasts. On the other hand, galectin-3 deficiency markedly reduces activated myofibroblast but does not affect macrophage recruitment nor pro-inflammatory cytokine profiles in injured tissue, such as IL-6 and TNF-α. Additionally, while galectin-3 deficiency led to reduced collagen deposition and reduced myofibroblast activation, TGF-β expression or smad2/3 phosphorylation were not influenced [19]. Furthermore, our unpublished galectin-3 chemotaxis assay results show that recombinant galectin-3 significantly induces monocyte migration, which could be markedly attenuated after treatment with galectin-3 inhibitors including modified citrus pectin (MCP) and lactose. These inhibitors act as a ligand, binding to galectin-3 s CRD. Altogether, these events clearly indicate galectin-3 to be a key player in the signal axis of fibrosis generation, specifically inducing macrophage and TGF-β induced myofibroblast activation [17, 19, 23, 42]. CLINICAL UTILITY OF GALECTIN-3 Clinical trials have consistently indicated a potential clinical utility of galectin-3 as a biomarker for prognosticating heart failure. Herein, van Kimmenade et al. were the first to evaluate the prognostic and predictive value of galectin-3 as a biomarker in acute heart failure [43], in the Pro-BNP investigation of dyspnea in the emergency department (PRIDE). While N-terminal pro brain natriuretic peptide (NT-proBNP) was a superior predictor for diagnosis of acute heart failure compared to galectin-3 and apelin (herein galectin-3 was a better predictor than apelin) Figure 2A, galectin-3 was the superior predictor compared to NTproBNP and apelin for prognosis in acute heart failure. Multivariate logistic regression analysis revealed that elevated plasma levels of galectin-3 were indeed the most powerful predictor for death, or the combination of death and recurrent heart failure within 60 days. Plasma galectin-3 levels were related to detailed echocardiographic examinations in substudy (N =115) of the PRIDE [44]. Galectin-3 levels were not strongly related to markers of LV structure or systolic function, but related to measures of RV function and diastolic dysfunction, and highest galectin-3 concentrations were strongly associated with a higher risk of 4 year mortality, is independent from LV dimensions, function, and RV pressure, supporting the role galectin-3 may play in fibrosis and progressive cardiac failure. 44

46 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure 3 Figure 2 Galectin-3 as a biomarker; data from several clinical trials. Figure 2A: Combined receiver-operating characteristic (ROC) curves for amino-terminal pro-brain natriuretic peptide (NT-proBNP), galectin-3 and apelin for the diagnosis of heart failure in dyspneic patients. The ROC analysis for NT-proBNP showed an area under the curve (AUC) for NT-proBNP of 0.94 (p = ). The ROC analysis for galectin-3 showed an AUC of 0.72 (p = ). The AUC for apelin for diagnosis of acute heart failure was 0.52 (p = 0.23). This figure was reprinted from van Kimmenade et al. [43], with permission. Figure 2B: Mortality as a function of baseline galectin-3 and NT-proBNP categories. The median value of NTproBNP (253 pmol/l), was used to define two levels of NT-proBNP concentration. Of the 232 subjects, 231 had both a galectin-3 and NT-proBNP measurement. The number of patients in the each category is as follows: high galectin-3 and high NT-proBNP (n=66); low galectin-3 and low NT-proBNP (n=69); low galectin-3 and high NT-proBNP (n=49); high galectin-3 and low NT-proBNP (n=47). Reprinted from reference [46] by Lok et al., with permission. Figure 2C: Graphical depiction of the risk estimates for experiencing the primary outcome in patients with HFPEF and HFREF with increasing levels of plasma galectin-3. The distribution of (log-transformed) galectin-3 is depicted in the background in brown bars. A similar increase in galectin-3 causes a much more pronounced increase in risk in patients with HFPEF compared to patients with HFREF. Figure reprinted from reference [47] by de Boer et al, with permission.figure 2D: Graph showing galectin-3 levels in male (blue line) and female subjects (red line) from the general population. Grey-shaded areas indicate 95% confidence intervals. Galectin-3 levels increase with increasing age, particularly in female subjects. This figure is reprinted from reference [49] by de Boer et al, with permission. 45

47 Chapter 3 Further and specifically relating galectin-3 to fibrosis, plasma galectin-3 levels were significantly correlated with several serum markers of cardiac ECM turnover, such as PIIINP, MMP-2 and TIMP-1, in 106 patients with chronic heart failure (New York Heart Association class II-III; mean LV ejection fraction [LVEF], 35+9%) [45]. Subsequently, Milting et al. described the kinetics of galectin-3 in 55 end stage heart failure patients with the need for mechanical circulatory support (MCS). Notably, this study found that fibrosis related biomarkers, such as tissue inhibitor of metalloproteinase (TIMP), tenascin C (TNC), OPN, BNP and galectin-3 were all increased in patients with terminal heart failure. Interestingly, MCS only reduced the loading related biomarker BNP, but none of the other fibrosis related biomarkers. Additionally, patients who did not survive on MCS had higher baseline galectin-3 levels when compared with patients who lived until transplantation [45]. Then, a larger study by Lok et al. comprising 232 patients with chronic heart failure (New York Heart Association function class III or IV) demonstrated that patients with high baseline levels of both galectin-3 and NT-proBNP had around 1.5- to 2-fold higher mortality rate Figure 2B [46]. This study demonstrated incremental value of galectin-3 over NT-proBNP alone. Additionally, a large study of 592 patients with heart failure (Coordinating study evaluating outcomes of Advising and Counseling in Heart failure, COACH trial, [47]), with mean follow-up of 18 months supported prognostic value of galectin-3 to predict rehospitalization and death after correction for age, gender, BNP, egfr and diabetes, but not after correction for LVEF. A subanalysis revealed that increased plasma galectin-3 levels represents a stronger incremental risk in patients with preserved LVEF (HFPEF) compared to the patients with reduced LVEF (HFREF) (P<0.001) even when absolute galectin-3 levels did not differ between patients with HFPEF and HFREF [47] Figure 2C. Furthermore, in the HF-ACTION study where plasma galectin-3 levels were assessed in 895 subjects with heart failure from a randomized, controlled trial of exercise training in patients with chronic heart failure with NYHA class II, III or IV symptoms, galectin-3 was associated with NYHA class, lower systolic blood pressure, higher creatinine, higher NTproBNP, and lower maximal oxygen consumption. However, this association diminished after adjustment for NT-proBNP [48]. Finally, recent data show that small increases in galectin-3 may confer increased CV risk in the general population, in subjects at risk for heart failure development. Briefly, 7968 subjects were included in this study from the Prevention of Renal and Vascular End-stage Disease (PREVEND) cohort (mean age of 50 ± 13 years, median follow-up of approximately 10 years). Plasma galectin-3 levels correlated very strong with age and sex Figure 2D, and weakly with a wide range of risk factors of CV disease, including blood pressure, serum lipids, body mass index, renal function and NT-proBNP. After correction for classical CV risk factors (smoking, blood pressure, cholesterol and diabetes), increased plasma galectin-3 levels independently predicted all-cause mortality in a large community/based cohort [49]. 46

48 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure Altogether, these available clinical studies have so far confirmed that plasma galectin-3 levels was significantly up regulated in acutely decompensated heart failure[43, 44, 50, 51], chronic heart failure [46-49] and end-stage heart failure with the need for mechanical circulatory support (MCS) [52]. Furthermore, clinical results from our group further demonstrated the predictive and prognostic value of galectin-3 [46, 47, 49]. A relationship between galectin-3 and cardiovascular (CV) risk factors was investigated in the general population of PREVEND study, and a strong gender specific interaction was revealed in the correlation between galectin-3 and cardiovascular risk factors [49]. Notably, the established link between plasma galectin-3 and fibrosis was also established in clinical biomarker studies [52] and needs to be explored further in order to develop more efficient strategies to target cardiac remodeling in heart failure leading to fibrosis. CONCLUSION AND MAPPING OF GALECTIN-3 PATHWAYS In conclusion, galectin-3 is highly expressed in the fibrotic area of the failing or stressed heart [14, 17, 19, 21, 22] and cardiac fibroblasts and macrophages are the main sources of galectin-3. Further, galectin-3 was shown to activate the TGF-β/Smad3 pathway [22], while 3 Figure 4: Working scheme representing the sequence of events following an index event (such as myocardial infarction (MI), hypertension, myocarditis and cardiomyopathy) leading to remodeling and non-remodeling heart failure. The graph within this graphic is taken from the de Boer et al. [8] and represents the adjusted Cox regression curves for quartiles of plasma galectin-3 showing the cumulative risk for the combined end-point, death. The back circles with the white numbers represent quartile 1 through 4, respectively. Galectin-3 is displayed as a central modulating factor involved in the remodeling process which leads to ongoing damage and eventually poor heart failure outcome. Therapeutically, galectin-3 inhibition could favor non-remodeling heart failure, thereby potentially improving heart failure outcome. Figure reprinted from reference [56] by de Boer et al. with permission. 47

49 Chapter 3 macrophages and the inflammatory factor TGF-β demonstrated to elevate galectin-3 expression. Then, galectin-3 inhibition or deficiency was found to not affect macrophage activation and TGF-β expression levels. Therefore, galectin-3 may be considered as an independent participant in macrophage and TGF-β/Smad modulation pathway. Other experimental animal studies reported that galectin-3 was not only significantly associated with myofibroblast induced collagen synthesis and deposition but was also markedly correlated with ECM fibrosis markers, such as: α-sma, COL1A1, COL3A1, TIMP, MMP [53] Figure 3(see Chapter 2 Figure 1). Lastly, relevant clinical studies identified that plasma galectin-3 is significantly correlated with serum extracellular fibrosis turn over biomarkers, like PINP, PIIINP, TIMP, MMP. Collectively, galectin-3 may be suggested as culprit biomarker involved in pathophysiology circle of cardiac remodeling and heart failure. A suggested pathway of galectin-3 is displayed in Figure 4 [56]. Data from experimental renal damage and cancer suggest to galectin-3 is a feasible target for therapy. Our pilot data lend support to the notion that also in heart failure galectin-3 may be a target for therapy. More research is warranted herein, specifically, at what stage galectin- 3 comes into play and what the ideal window would be for intervention. Most data show that preventative regimen might work, but if fibrosis, once ensued, could be attenuated or reversed is also unknown. From a clinical point of view this is of utmost importance. Disclosures BG Medicine Inc, (BGM, Waltham, MA, USA) hold certain rights with respect to the use of galectin-3 in heart failure. The UMCG, department of Cardiology, which employs the authors, received research grants from BGM. Dr. de Boer received consultancy fees from BGM. Acknowledgement We thank Maxi Meissner, PhD for her assistance in preparing this manuscript. 48

50 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure References 1. Seetharaman J, Kanigsberg A, Slaaby R, et al. X-ray crystal structure of the human galectin-3 carbohydrate recognition domain at 2.1-A resolution. J Biol Chem 273: , Henrick K, Bawumia S, Barboni EA, et al. Evidence for subsites in the galectins involved in sugar binding at the nonreducing end of the central galactose of oligosaccharide ligands: sequence analysis, homology modeling and mutagenesis studies of hamster galectin-3. Glycobiology 8:45-57, Rosenberg I, Cherayil BJ, Isselbacher KJ, et al. Mac-2-binding glycoproteins. Putative ligands for a cytosolic betagalactoside lectin. J Biol Chem 266: , Sato S, Hughes RC. Binding specificity of a baby hamster kidney lectin for H type I and II chains, polylactosamine glycans, and appropriately glycosylated forms of laminin and fibronectin. J Biol Chem 267: , Ochieng J, Furtak V, Lukyanov P. Extracellular functions of galectin-3. Glycoconj J 19: , de Boer RA, Yu L, van Veldhuisen DJ. Galectin-3 in cardiac remodeling and heart failure. Curr Heart Fail Rep 7:1-8, Hughes RC. The galectin family of mammalian carbohydrate-binding molecules. Biochem Soc Trans 25: , Hughes RC. Secretion of the galectin family of mammalian carbohydrate-binding proteins. Biochim Biophys Acta 1473: , Woo HJ, Shaw LM, Messier JM, et al. The major non-integrin laminin binding protein of macrophages is identical to carbohydrate binding protein 35 (Mac-2). J Biol Chem 265: , Koths K, Taylor E, Halenbeck R, et al. Cloning and characterization of a human Mac-2-binding protein, a new member of the superfamily defined by the macrophage scavenger receptor cysteine-rich domain. J Biol Chem 268: , Probstmeier R, Montag D, Schachner M. Galectin-3, a beta-galactoside-binding animal lectin, binds to neural recognition molecules. J Neurochem 64: , Ochieng J, Warfield P. Galectin-3 binding potentials of mouse tumor EHS and human placental laminins. Biochem Biophys Res Commun 217: , Kim H, Lee J, Hyun JW, et al. Expression and immunohistochemical localization of galectin-3 in various mouse tissues. Cell Biol Int 31: , Sharma UC, Pokharel S, van Brakel TJ, et al. Galectin-3 marks activated macrophages in failure-prone hypertrophied hearts and contributes to cardiac dysfunction. Circulation 110: , Kasper M, Hughes RC. Immunocytochemical evidence for a modulation of galectin 3 (Mac-2), a carbohydrate binding protein, in pulmonary fibrosis. J Pathol 179: , Wang L, Friess H, Zhu Z, et al. Galectin-1 and galectin-3 in chronic pancreatitis. Lab Invest 80: , Henderson NC, Mackinnon AC, Farnworth SL, et al. Galectin-3 regulates myofibroblast activation and hepatic fibrosis. Proc Natl Acad Sci U S A 103: , Sasaki S, Bao Q, Hughes RC. Galectin-3 modulates rat mesangial cell proliferation and matrix synthesis during experimental glomerulonephritis induced by anti-thy1.1 antibodies. J Pathol 187: , Henderson NC, Mackinnon AC, Farnworth SL, et al. Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol 172: ,

51 Chapter Eis V, Luckow B, Vielhauer V, et al. Chemokine receptor CCR1 but not CCR5 mediates leukocyte recruitment and subsequent renal fibrosis after unilateral ureteral obstruction. J Am Soc Nephrol 15: , Sharma U, Rhaleb NE, Pokharel S, et al. Novel anti-inflammatory mechanisms of N-Acetyl-Ser-Asp-Lys-Pro in hypertension-induced target organ damage. Am J Physiol Heart Circ Physiol 294:H , Liu YH, D'Ambrosio M, Liao TD, et al. N-acetyl-seryl-aspartyl-lysyl-proline prevents cardiac remodeling and dysfunction induced by galectin-3, a mammalian adhesion/growth-regulatory lectin. Am J Physiol Heart Circ Physiol 296:H404-12, Mackinnon AC, Gibbons MA, Farnworth SL, et al. Regulation of Transforming Growth Factor-beta1-driven Lung Fibrosis by Galectin-3. Am J Respir Crit Care Med 185: , Takenaka Y, Fukumori T, Yoshii T, et al. Nuclear export of phosphorylated galectin-3 regulates its antiapoptotic activity in response to chemotherapeutic drugs. Mol Cell Biol 24: , Sato S, Burdett I, Hughes RC. Secretion of the baby hamster kidney 30-kDa galactose-binding lectin from polarized and nonpolarized cells: a pathway independent of the endoplasmic reticulum-golgi complex. Exp Cell Res 207:8-18, Sato S, Hughes RC. Regulation of secretion and surface expression of Mac-2, a galactoside-binding protein of macrophages. J Biol Chem 269: , Nickel W. The mystery of nonclassical protein secretion. A current view on cargo proteins and potential export routes. Eur J Biochem 270: , Jiang JX, Chen X, Hsu DK, et al. Galectin-3 modulates phagocytosis-induced stellate cell activation and liver fibrosis in vivo. Am J Physiol Gastrointest Liver Physiol 302:G439-46, de Oliveira SA, de Freitas Souza BS, Sa Barreto EP, et al. Reduction of galectin-3 expression and liver fibrosis after cell therapy in a mouse model of cirrhosis. Cytotherapy 14: , Kolatsi-Joannou M, Price KL, Winyard PJ, et al. Modified citrus pectin reduces galectin-3 expression and disease severity in experimental acute kidney injury. PLoS One 6:e18683, Thandavarayan RA, Watanabe K, Ma M, et al protein regulates Ask1 signaling and protects against diabetic cardiomyopathy. Biochem Pharmacol 75: , Kamal FA, Watanabe K, Ma M, et al. A novel phenylpyridazinone, T-3999, reduces the progression of autoimmune myocarditis to dilated cardiomyopathy. Heart Vessels 26:81-90, Psarras S, Mavroidis M, Sanoudou D, et al. Regulation of adverse remodelling by osteopontin in a genetic heart failure model. Eur Heart J, Wilker E, Yaffe MB Proteins--a focus on cancer and human disease. J Mol Cell Cardiol 37: , Banerjee A, Apte UM, Smith R, et al. Higher neutrophil infiltration mediated by osteopontin is a likely contributing factor to the increased susceptibility of females to alcoholic liver disease. J Pathol 208: , Fisher AD. 'Maxillofacial surgery should become a specialty of medicine'. Br Dent J 172:46-47, Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 275: , Brown RD, Ambler SK, Mitchell MD, et al. The cardiac fibroblast: therapeutic target in myocardial remodeling and failure. Annu Rev Pharmacol Toxicol 45: ,

52 Role of Galectin-3 Pathways in the Pathogenesis of Cardiac Remodeling and Heart Failure 39. de Cavanagh EM, Ferder M, Inserra F, et al. Angiotensin II, mitochondria, cytoskeletal, and extracellular matrix connections: an integrating viewpoint. Am J Physiol Heart Circ Physiol 296:H550-8, Krzeslak A, Lipinska A. Galectin-3 as a multifunctional protein. Cell Mol Biol Lett 9: , Dvorankova B, Szabo P, Lacina L, et al. Human galectins induce conversion of dermal fibroblasts into myofibroblasts and production of extracellular matrix: potential application in tissue engineering and wound repair. Cells Tissues Organs 194: , Henderson NC, Sethi T. The regulation of inflammation by galectin-3. Immunol Rev 230: , van Kimmenade RR, Januzzi JL,Jr, Ellinor PT, et al. Utility of amino-terminal pro-brain natriuretic peptide, galectin-3, and apelin for the evaluation of patients with acute heart failure. J Am Coll Cardiol 48: , Shah RV, Chen-Tournoux AA, Picard MH, et al. Galectin-3, cardiac structure and function, and long-term mortality in patients with acutely decompensated heart failure. Eur J Heart Fail 12: , Lin YH, Lin LY, Wu YW, et al. The relationship between serum galectin-3 and serum markers of cardiac extracellular matrix turnover in heart failure patients. Clin Chim Acta 409:96-99, Lok DJ, Van Der Meer P, de la Porte PW, et al. Prognostic value of galectin-3, a novel marker of fibrosis, in patients with chronic heart failure: data from the DEAL-HF study. Clin Res Cardiol 99: , de Boer RA, Lok DJ, Jaarsma T, et al. Predictive value of plasma galectin-3 levels in heart failure with reduced and preserved ejection fraction. Ann Med 43:60-68, Felker GM, Fiuzat M, Shaw LK, et al. Galectin-3 in ambulatory patients with heart failure: results from the HF-ACTION study. Circ Heart Fail 5:72-78, de Boer RA, van Veldhuisen DJ, Gansevoort RT, et al. The fibrosis marker galectin-3 and outcome in the general population. J Intern Med 272:55-64, Christenson RH, Duh SH, Wu AH, et al. Multi-center determination of galectin-3 assay performance characteristics: Anatomy of a novel assay for use in heart failure. Clin Biochem 43: , Grandin EW, Jarolim P, Murphy SA, et al. Galectin-3 and the development of heart failure after acute coronary syndrome: pilot experience from PROVE IT-TIMI 22. Clin Chem 58: , Kramer F, Milting H. Novel biomarkers in human terminal heart failure and under mechanical circulatory support. Biomarkers 16 Suppl 1:S31-41, de Boer RA, Voors AA, Muntendam P, et al. Galectin-3: a novel mediator of heart failure development and progression. Eur J Heart Fail 11: , Yu L, Ruifrok WP, Sillje HHW, et al. Genetic disruption of galectin-3 prevents adverse cardiac remodeling. European Heart Journal; 32 (Abstract Supplement): 1094, Yu L, Ruifrok WP, Sillje HHW, et al. Pharmacological inhibition of galectin-3 attenuates cardiac remodeling and heart failure. European Heart Journal; 32 (Abstract Supplement): 1097, de Boer RA, Meissner M, van Veldhuisen DJ. Chapter 13: Galectin-3. In: Alan S Maisel (ed) Cardiac Biomarkers: Expert Advice for Clinicians, 1st edition. New Delhi, Panama city, London, pp , ISBN

53

54

55

56 Chapter 4 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling by Interfering with Myocardial Fibrogenesis Lili Yu ; Willem P.T. Ruifrok ; Maxi Meissner; Eelke M. Bos ; Harry van Goor; Bahram Sanjabi; Pim van der Harst; Bertram Pitt; Irwin J. Goldstein; Jasper A. Koerts; Dirk J. van Veldhuisen; Ruud A. Bank; Wiek H. van Gilst; Herman H.W. Silljé; Rudolf A. de Boer Provisionally accepted in Circulation (heart failure)

57 Chapter 4 Abstract Background - Galectin-3 has been implicated in the development of organ fibrosis. It is unknown whether it is a relevant therapeutic target in cardiac remodeling and heart failure (HF). Methods and Results - Galectin-3 knock-out (Gal3-KO) and wild-type (WT) mice were subjected to angiotensin II (AngII) infusion (2.5 µg/kg for 14 days) or transverse aortic constriction (TAC) for 28 days to provoke cardiac remodeling. The efficacy of the galectin-3 inhibitor N-acetyllactosamine (Gal3i) was evaluated in TGR(mREN2)27 (REN2) rats and in WT mice with the aim of reversing established cardiac remodeling following TAC. In WT mice, AngII and TAC perturbations caused left ventricular (LV) hypertrophy, decreased fractional shortening and increased LV end-diastolic pressure and fibrosis (P<0.05 vs. control WT). Gal3-KO mice also developed LV hypertrophy but without LV dysfunction and fibrosis (P=NS). In REN2 rats, pharmacological inhibition of galectin-3 attenuated LV dysfunction and fibrosis. To elucidate the beneficial effects of galectin-3 inhibition on myocardial fibrogenesis, cultured fibroblasts were treated with galectin-3 in the absence or presence of Gal3i. Inhibition of galectin-3 was associated with a downregulation in collagen production (collagen I and III), collagen processing, cleavage, cross-linking and deposition. Similar results were observed in REN2 rats. Inhibition of galectin-3 also attenuated the progression of cardiac remodeling in a long-term TAC mouse model. Conclusions - Genetic disruption and pharmacological inhibition of galectin-3 attenuates cardiac fibrosis, LV dysfunction and subsequent HF development. Drugs binding to galectin-3 may be potential therapeutic candidates for the prevention or reversal of HF with extensive fibrosis. Keywords Galectin-3, cardiac remodeling, heart failure, fibrosis, renin-angiotensin system 56

58 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Introduction Galectin-3 belongs to the galectin family of mammalian lectins and is characterized by a carbohydrate recognition domain (CRD) that has affinity for β-galactosides. Galectin-3 mediates cell cell and cell matrix interactions by binding to lactosamine-containing cell surface glycoconjugates via its CRD. There is mounting evidence demonstrating its key role in inflammatory and fibrotic processes. 1 In experimental models, galectin-3 was found to be involved in the development of liver and kidney fibrosis. 2,3 It has also been associated with cardiac fibrosis in TGRmREN2-27 (REN2) rats with experimental heart failure (HF). 4 A continuous intrapericardial infusion of low-dose galectin-3 resulted in cardiac fibrosis and left ventricular (LV) dysfunction in both failure-prone hypertrophic REN2 and healthy Sprague-Dawley rats. 4,5 The upregulation of galectin-3 has also been demonstrated in failing human hearts 4 and circulating levels are a powerful predictor of outcome in both acute and chronic HF. This observation has led to the approval of galectin-3 by the Food and Drug Administration as a novel biomarker in HF that may help categorize patients in the remodeling and non-remodeling stages of HF. 6,7 Despite evidence suggesting the involvement of galectin-3 in the pathophysiology of HF, it remains unknown if it contributes actively to the disease. In the present study, we determined if the disruption of the galectin-3 gene, shown to attenuate hepatic and renal fibrosis in galectin-3 knock-out (Gal3-KO) mice, 2,8 would prevent the development of HF in animals with active cardiac remodeling. Furthermore, because the beneficial effects of targeted CRD inhibition of galectin-3 have been demonstrated in a model of kidney injury, 3 we also determined the effects of a galectin-3 inhibitor (Gal3i), N-acetyllactosamine (N-Lac), which has a high affinity for galectin-3 CRD, in TGR(mREN2)27 (REN2) rats with HF and in mice subjected to transverse aortic constriction (TAC). 4 Methods For detailed description of the methods, please refer to the Online Supplemental Data. Animals Male Gal3-KO mice were generated and bred at the Jackson Laboratory (Bar Harbor, ME, USA) as described previously. 2,8,9 Age-matched male transgene-negative wild-type (WT) littermates were used as controls. Supplemental Figure S1A depicts the generation of Gal3- KO mice. Male homozygous REN2 (Max Delbrück Center for Molecular Medicine, Berlin Buch, Germany) and age-matched male Sprague-Dawley (SD) rats (controls) were used as described previously All experiments were approved by the Animal Ethical Committee of the University of Groningen (the Netherlands) and conducted in accordance with existing guidelines on the care and use of laboratory animals. Mouse experiments 57

59 Chapter 4 Six- to 10-week old Gal3-KO and WT mice were subjected to an infusion of angiotensin II (AngII) (2.5 µg/kg/day) for 14 days or LV pressure overload by TAC for 28 days (prevention experiment). 13 In another series of experiments (reversal experiment), six- to eight-week old male C56Bl6/J mice (Harlan, The Netherlands) underwent TAC for 28 days and were then treated with N-Lac (5 mg/kg/day, intraperitoneal (i.p.) injections) three times a week for 28 days (starting day 28 until day 56). 14 Rat experiments SD and REN2 rats were treated with N-Lac (5 mg/kg/day, i.p.) three times a week for six weeks. Other experiments Cardiac function was studied with echocardiography and hemodynamic measurement as described previously. 15,16 Immunohistochemical analyses were performed and collagen digestibility was determined. 17 Cell cultures of human adult dermal fibroblasts were used to study galectin-3-mediated fibrogenesis and the effects of N-Lac. Cardiac and fibroblast gene expression was measured using RT-qPCR as described previously. 18 The lists of primers are presented in Supplemental Tables S1, 2 and 3. Statistical analyses Results are reported as mean ± standard error of the mean (SEM). Mice were analyzed in two separate subgroups comparing genotype differences (WT vs. KO) and model differences (control vs. AngII or sham vs. TAC). Subgroup A consisted of WT-control, Gal3-KO-control, WT-AngII, Gal3-KO-AngII and subgroup B consisted of WT-sham, Gal3-KO-sham, WT- TAC and Gal3-KO-TAC. Levene s test was used to test homogeneity of variances within parameters. If there was equality of variances, statistical analyses were performed by one-way ANOVA with Bonferroni post-hoc tests (mice group A M=6 tests, mice group B M=6 tests, rats M=3 tests). If there was unequality of variances, statistical analyses were performed by Welch s ANOVA with Games-Howell post-hoc test. Cell experiments were analyzed with Kruskall-Wallis test (N=3 per group). For the TAC reversal experiment, differences between saline treated mice and N-Lac treated mice were analyzed at the eight week time point using an unpaired t-test (N=7-9). Baseline TAC reversal is depicted as a dotted line. In all figures, only relevant comparisons are shown by the symbols for reasons of clarity. All P-values are two-tailed and a value less than 0.05 was considered significant. All analyses were performed using SPSS version 20.0 software (SPSS, Chicago, IL, USA). Results Galectin-3 knock-out mice are protected against left ventricular dysfunction 58

60 Table 1. Baseline characteristics and hemodynamic data at sacrifice of mice Gal3-KO-TAC (N=14) WT-TAC (N=12) Gal3-KOsham (N=11) WT-sham (N=7) Gal3-KO- AngII (N=10) WT-AngII (N=12) Gal3-KO-con (N=10) WT-con (N=10) BW (g) 28.5± ± ±0.7 * 24.3± ± ± ± ±0.4 LV weight 5.90± ± ±0.5 * 8.28± ± ± ± ±0.8 MAP (mmhg) 71±2 67±2 97±1 * 97±2 75±2 77±2 82±6 79±7 IVSd (mm) 0.64± ± ±0.03 * 0.85± ± ± ± ±0.02 # LVEDD (mm) 3.78± ± ± ± ± ± ± ±0.04 LVESD (mm) 2.09± ± ±0.12 * 2.15± ± ± ± ± 0.09 **# LVPWd (mm) 0.65± ± ±0.02 * 0.91± ± ± ± ±0.02 # Corrected dpdtmax 102±1 94±7 74±6 * 89±5 103±6 90±7 60±4 64±2 AngII: angiotensin II; TAC: transverse aortic constriction; BW: body weight; LV: left ventricular; LV weight: corrected for tibia length (mg/mm), MAP: mean arterial pressure; IVSd: thickness of the interventricular septum in diastole; LVEDD: left ventricular end diastolic diameter; LVESD: left ventricular end systolic diameter; LVPWd: thickness of the left ventricular posterior wall in diastole; dpdtmax is an index of maximal contraction of the left ventricle and is corrected for peak systolic pressure. * P<0.05 vs. WT-con, P<0.05 vs. Gal3-KO-con, P<0.05 vs. WT-AngII. P<0.05 vs. WT-sham, # P<0.05 vs. Gal3-KO-sham, ** P<0.05 vs. WT-TAC

61 Chapter 4 Figure 1 Hemodynamic data in mice and rats at sacrifice. (a) Outline of the experimental protocol of the prevention studies in mice and rats. (b) Expression of atrial natriuretic peptide (ANP) mrna in mouse hearts. (c) Fractional shortening in mouse hearts assessed with echocardiography. (d) Left ventricular end diastolic pressure (LVEDP) in mouse hearts. (e) Isovolumetric relaxation constant Tau in mouse hearts. (f) dpdtmin corrected for peak systolic pressure in mouse hearts. (g) Expression of ANP mrna in rat hearts. (h) Fractional shortening in rat hearts assessed with echocardiography. (i) Change in fractional shortening in rat hearts assessed with echocardiography at baseline and prior to sacrifice. (j) LVEDP in rat hearts. (k) Tau in rat hearts. AngII: angiotensin II; TAC: transverse aortic constriction; Wk: week; Gal3i: galectin-3 inhibitor (N-Lac: N- acetyllactosamine); con: control. N=5-12 per group. * P<0.05 vs. WT-con, P<0.05 vs. Gal3-KO-con, P<0.05 vs. WT-AngII, P<0.05 vs. WT-sham, # P<0.05 vs. Gal3-KO-sham, ** P<0.05 vs. WT-TAC, P<0.05 SDcon vs. REN2-con, P<0.05 REN2-con vs. REN2-Gal3i, P<0.05 SD-con vs. REN2-Gal3i, ## P<0.05 vs. all other groups at sacrifice. 60

62 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling The intervention and treatment schemes of the WT and Gal3-KO mice are presented in Figure 1a and the baseline characteristics and hemodynamic data at sacrifice are presented in Table 1. When compared with control or sham, LV galectin-3 expression was increased almost 2-fold in WT mice treated with AngII or TAC while galectin-3 expression was absent in Gal3-KO mice (mrna and protein, Supplemental Figure S1 b-d). In the WT mice, both interventions caused LV hypertrophy as evidenced by the increases in LV weight (normalized to tibia length, Table 1), wall thicknesses (Table 1) and LV atrial natriuretic peptide (ANP) expression (Figure 1b) along with a decrease in contractile function (fractional shortening, FS, Figure 1c). Hemodynamic measurements revealed LV relaxation impairment in the WT groups (Figure 1d-f) with increases in LVEDP and Tau (an isovolumetric relaxation constant measured according to the Glantz method) along with decreases in dpdtmin (corrected for peak systolic pressure). As shown in Table 1, both Gal3-KO and WT mice subjected to AngII infusion or TAC had a similar degree of LV hypertrophy. However, and irrespective of the perturbation, Gal3- KO mice had preserved FS (Figure 1c). Hemodynamic measurements revealed that Gal3-KO mice were protected against LV relaxation impairment following AngII infusion or TAC (Figures 1d-f), which did not result in changes in LVEDP and Tau (P=NS vs. respective controls). The only exception was the corrected dpdtmin which was significantly decreased in the Gal3-KO-AngII group. Inhibition of galectin-3 with N-acetyllactosamine prevents left ventricular dysfunction in failure-prone REN2 rats The intervention and treatment schemes of the SD and REN2 rats are presented in Figure 1a. Table 2 shows the baseline characteristics and hemodynamic data at sacrifice. As expected, LV weight (adjusted for tibia length) was significantly increased in the untreated REN2 rats. Treatment with N-Lac did not prevent the development of LV hypertrophy or decrease LV ANP levels (Table 2, Figure 1g). Fractional shortening (FS) progressively declined in the untreated REN2 rats (Figure 1i) but was preserved in the Gal3i-treated rats (Figure 1h). Hemodynamic measurements revealed an increased LVEDP in the untreated REN2 rats as compared with the SD rats (Figure 1j) as well as Tau (Figure 1k), associated with increased lung weights (Table 2), all suggestive of developing HF. Treatment with Gal3i reduced LVEDP in REN2, and lung weigh, but not Tau. Finally, accelerated cardiac remodeling in untreated REN2 rats was associated with poorer survival than Gal3i-treated REN2 rats (Supplemental Figure S2). 4 Galectin-3 disruption or inhibition attenuates the formation of fibrosis in the heart To determine if galectin-3 is actively involved in the formation of fibrosis, we analyzed the presence of myocardial fibrosis. Figures 2a and 2b show representative pictures of fibrotic tissue and the fibrosis score in mouse hearts. The hearts of control and sham-operated WT mice had very little fibrosis ( 2%). A significantly higher percentage of fibrosis was evident 61

63 Table 2. Baseline characteristics and hemodynamic data at sacrifice of rats SD-con (N=12) REN2-con (N=18) REN2-Gal3i (N=10) BW (g) 362±11 304±7 374±29 LV weight 21.9± ± ±1.3, Lung weight 4.4± ± ±0.2 MAP (mmhg) 97±5 77±6 77±3 IVSd (mm) 1.66± ± ±0.02 LVPWd (mm) 1.73± ± ±0.05 LVEDD (mm) 6.15± ± ±0.32 LVESD (mm) 3.61± ± ±0.19 Corrected dpdtmax 65±5 58±2 70±2 Corrected dpdtmin -79±6-57±2-62±4 Gal3i: galectin-3 inhibitor; BW: body weight; LV: left ventricular; LV weight: corrected for tibia length (mg/mm); Lung weight: corrected for BW, mg/gr; MAP: mean arterial pressure; IVSd: thickness of the interventricular septum in diastole; LVPWd: thickness of the left ventricular posterior wall in diastole; LVEDD: left ventricular end diastolic diameter; LVESD: left ventricular end systolic diameter; dpdtmax and dpdtmin are indices of maximal contraction and relaxation of the left ventricle, they are corrected for peak systolic pressure. P<0.05 SD-con vs. REN2-con, P<0.05 REN2-con vs. REN2-Gal3i, P<0.05 SD-con vs. REN2-Gal3i

64 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Figure 2 Fibrosis in mouse and rat hearts. (a) Typical examples of fibrosis staining with Masson s trichrome staining in mouse hearts. (b) Percentage of fibrosis in mouse hearts. (c) Typical examples of fibrosis staining in rat hearts. (d) Percentage of fibrosis in rat hearts. (e) Collagen cross-linking represented as collagen digestibility in rat hearts. (f) Level of PINP measured in rat blood plasma. Con: control; AngII: angiotensin II; TAC: transverse aortic constriction; Gal3i: galectin-3 inhibitor. N=5-12 per group. * P<0.05 vs. WT-con, P<0.05 vs. WT-AngII, P<0.05 vs. WT-sham, ** P<0.05 vs. WT-TAC, P<0.05 SD-con vs. REN2-con, P<0.05 REN2-con vs. REN2-Gal3i, P<0.05 SD-con vs. REN2-Gal3i 4 in the WT-AngII and WT-TAC mice (Figure 2b). However, neither AngII infusion nor TAC resulted in increased fibrosis in Gal3-KO animals. Similar results were also observed in rat hearts. Compared with SD rats, REN2 rats exhibited a high percentage of fibrotic tissue and treatment with Gal3i significantly reduced the percentage of fibrosis in the REN2 rats (Figure 2c, 2d). We evaluated collagen digestibility (% collagen released by proteolytic enzymes) as a measure of the extent of collagen cross-linking. In our assay, the higher the numbers of crosslinks, the lower the amount of released collagen. 19,20 Compared with SD rats, collagen digestibility was significantly reduced in the hearts of REN2 rats (Figure 2e). As collagen fibers in fibrotic lesions display a higher level of pyridinoline cross-links making them more resistant to the enzymatic action of proteinases, 20 our results indicate the presence of more cross-linked collagen in the hearts from untreated REN2 rats (Figure 2e). Treatment of REN2 rats with Gal3i resulted in a collagen digestibility similar to that of SD rats (figure 2e). We also measured the plasma levels of PINP, a marker of collagen cleavage, with an ELISA assay. In REN2 rats, PINP was significantly increased compared with SD rats and inhibition of galectin-3 resulted in a significant reduction of PINP concentration (Figure 2f). Collectively, these results suggest that Gal3i can reduce pathological fibrosis through a reduction of collagen deposition and synthesis along with increased collagen digestibility. 63

65 Chapter 4 Figure 3 Histological analysis of rat hearts. (a) Typical examples of Gomori staining for cardiomyocyte size, αsma staining for myofibroblasts, CD68 staining for macrophages and PCNA staining for proliferating cells. (b) Quantification of cardiomyocyte size. (c) Quantification of α-sma staining. (d) Quantification of macrophages per mm2. (e) Quantification of proliferating cells per mm2. Con: control; Gal3i: galectin-3 inhibitor. N=5-8 per group. P<0.05 SD-con vs. REN2-con, P<0.05 REN2-con vs. REN2-Gal3i, P<0.05 SD-con vs. REN2Gal3i 64

66 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Galectin-3 inhibition reduces the number of α-sma positive cells in the heart Figure 3 shows typical examples of Gomori, α-sma, CD68 (a macrophage marker) and PCNA staining in rat hearts with their respective quantification. Cardiomyocyte size and α- SMA expression were significantly increased in REN2 rats. Treatment with Gal3i did not decrease cardiomyocyte size (Figure 3a, 3b) but it normalized α-sma expression indicating a lower number of myofibroblasts in these groups (Figure 3a, 3c). The number of macrophages was also increased in both REN2 groups (Figure 3a, 3d), as was the number of proliferating cells, although this was only significant in the untreated REN2 group (Figure 3a, 3e). Immunohistochemical analysis revealed that galectin-3 immunoreactivity was predominantly observed in the interstitial space, but not in cardiomyocytes (Supplemental figure S3). In an effort to determine the localization of cardiac galectin-3 and its source, we conducted further studies. We observed that galectin-3 co-localized with macrophages (Supplemental figure S4 a-d) and at sites of collagen deposition (Supplemental figure S4, e-h). Furthermore, galectin-3 4 Figure 4 Effects of galectin-3 and Gal3i on human dermal fibroblasts (HDF) (a) Typical examples of HDF appearance with or without treatment. (b) Confirmation of fibroblast phenotype by vimentin staining. (c) Most significantly regulated genes from the Low Density Array. (d-j) confirmation of differential expression of genes identified in the low density array by RT-qPCR: (d) HSP47, (e) COL1A1, (f) COL1A2, (g) COL3A1, (h) P4HB, (i) PCOLCE, (j) LOXL2. (k) Amount of collagen 3 (COL3) in the tissue culture medium measured with ELISA after 72h treatment. CM: culture medium; Gal3: galectin-3; N-Lac: N-acetyllactosamine. N=3 per group. 65

67 Chapter 4 protein expression was highest in macrophages. Galectin-3 protein expression was also clearly detectable in fibroblast, but was not detectable in cardiomyocytes (Supplemental figure S5). Galectin-3 inhibition prevents galectin-3-mediated effects in human dermal fibroblasts Stimulation of human dermal fibroblasts with recombinant galectin-3, with or without Gal3i, did not lead to visible differences in fibroblast appearance (Figure 4a). Fibroblast phenotype was confirmed by staining for the mesenchymal marker vimentin (Figure 4b). Since our preliminary results demonstrated that the mrna level of COL1A1 in the presence of galectin-3 peaked 72 hours post-stimulation (data not shown), the expression of various fibrillar collagens and proteins involved in their processing was measured at the 72h time Figure 5 Changes in gene expression of fibrotic genes in rat hearts. (a-h) mrna expression in rat hearts: (a) Galectin-3, (b) Hsp47, (c) Col1a1, (d) Col1a2, (e) Col3a1, (f) P4hb, (g) Pcolce, (h) Loxl2, N=5-12 per group. (i) Western Blots for galectin-3, Hsp47 and GAPDH. (j) Galectin-3 protein expression measured with Western Blot. (k) Hsp47 protein expression measured with Western Blot. N=4 per group. Gal3i: galectin-3 inhibitor. P<0.05 SD-con vs. REN2-con, P<0.05 REN2-con vs. REN2-Gal3i, P<0.05 SD-con vs. REN2-Gal3i. 66

68 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling point with a low density array (overview in Supplemental Table S4; results are shown in Figure 4). The expression of COL1A1, COL1A2, COL3A1 and the collagen-modifying proteins encoded by P4HB, HSP47, PCOLCE and LOXL2 were all upregulated by galectin-3 (Figure 4, d-j) and inhibited by Gal3i co-treatment. The modulation of COL3A1 transcript expression by galectin-3 and Gal3i was also reflected in collagen type III levels in the culture medium measured with an ELISA (Figure 4 k). Genes involved in myocardial fibrogenesis The expression of the above-mentioned collagens and fibrosis-associated proteins was subsequently analyzed in the LV tissue of SD and REN2 rats by RT-qPCR (figure 5, a-h; primers are listed in Supplemental Table S2). These results confirm our in vitro observations with human fibroblasts and show that Gal3i reduces the expression of pro-fibrotic genes also 4 Figure 6 Gal3i inhibits progression of cardiac remodeling in mice. (a) Outline of the experimental protocol. (b) Left ventricular (LV) weight corrected for tibia length (TL) at sacrifice. (c) Fractional shortening assessed with echocardiography at sacrifice. (d) Left ventricular end diastolic pressure (LVEDP). (e) Atrial natriuretic peptide (ANP) mrna expression in mouse hearts. (f) Amount of fibrosis in mouse hearts. (g) Typical examples of Masson s trichome staining in mouse hearts. TAC: transverse aortic constriction; Wk(s): week(s); Gal3i: galectin-3 inhibitor; sal: saline, NS: no-significant. N=7-9 per group. Differences between groups were compared at the eight weeks timepoint (t-test), baseline four weeks is depicted as a dotted line. 67

69 Chapter 4 in vivo. We also measured the LV protein expression of galectin-3 and Hsp47 in lysates from SD and REN2 rat hearts (figure 5, i-k). Galectin-3 and Hsp47 protein levels were increased in untreated REN2 rats. Treatment with the Gal3i resulted in decreased Hsp47 (figure 5k) but not galectin-3 (figure 5j) protein levels, which reflects the changes in mrna expression (figure 5a-b). Finally, we analyzed the transcript levels of matrix metalloproteinases (MMP) 2 and 9 and tissue inhibitors of matrix proteases (TIMP) 1 and 2 in LV tissue of SD and REN2 rats. Mmp-9, Timp-1, and Timp-2 levels were increased in untreated REN2 rats and reduced by treatment with a Gal3i (Supplemental figure S6). Galectin-3 inhibition prevents further progression of established left ventricular remodeling We provoked LV remodeling with TAC surgery followed by an observational period of 28 days without intervention (Figure 6a). We then treated the mice with N-Lac or saline for another 28 days. At the end of the follow-up period, no increase in LV weight (figure 6b) was observed. Treatment with Gal3i for 28 days had no effect on LV weight, compared with the LV weights after the 28-day observational period (Figure 6b). However, a gradual progression of LV remodeling was observed in the untreated mice as evidence by the further decline in FS (Figure 6c) and by an increase in ANP levels (Figure 6e) and fibrosis (Figure 6f, g). In the Gal3i-treated mice, LVEDP was lower compared to untreated mice (Figure 6d). Discussion The current study provides several lines of evidence that galectin-3 is an active contributor in the development of cardiac remodeling, myocardial fibrogenesis and HF. We have demonstrated that inhibition of galectin-3 function by genetic disruption or pharmacological intervention halts the progression of cardiac remodeling, attenuates myocardial fibrogenesis and preserves LV function. These beneficial effects can be explained, at least in part, by the lower number of myofibroblasts in combination with diminished collagen synthesis, processing and cross-linking. Collectively, our results suggest that galectin-3 may be an attractive target for the prevention and treatment of HF. Disruption of galectin-3 attenuates cardiac remodeling and preserves cardiac function To explore the hypothesis that galectin-3-targeted interventions may protect against progressive cardiac remodeling and dysfunction, two experimental approaches were used in well-established mouse and rat models of cardiac remodeling: 1) complete genetic disruption of galectin-3 and 2) pharmacological inhibition with an agent that specifically binds to the CRD of galectin-3. Both the genetic disruption and pharmacological inhibition of galectin-3 resulted in considerable attenuation of cardiac remodeling and, specifically, to an almost it is complete inhibition of cardiac fibrosis. Functionally, the inhibition of galectin-3 improved diastolic dysfunction to a large extent (less increase in end-diastolic LV pressure and improved LV relaxation) in spite of the presence of significant LV hypertrophy. In this respect, 68

70 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling it is noteworthy that elevated levels of circulating galectin-3 have been shown to be strong predictors of poor outcome in patients with diastolic HF or HF with preserved LV ejection fraction. 21 Although the animals in our experimental models were mainly characterized by impaired diastolic function, mild (AngII, TAC-prevention) to moderate (REN2, TAC-reversal) systolic dysfunction also developed over time. Inhibition of galectin-3 preserved systolic function, a finding that was above all apparent in the reversal experiment where treatment was started after four weeks of TAC. In the untreated mice, mild systolic dysfunction was present after four weeks (figure 6c) and progressed over another four weeks of follow-up to overt systolic dysfunction (figure 6c). However, when mice were treated with the galectin-3 inhibitor, progression of systolic dysfunction was attenuated. These results suggest that galectin-3 inhibition might afford functional protection against developing and progressive cardiac remodeling. Additional evidence of the important role of galectin-3 in cardiac remodeling was obtained by treating REN2 rats with N-acetyllactosamine, an established inhibitor that binds to the galectin-3 CRD. 14,22,23 The homozygous REN2 rat model is a well described model of rapidly progressive cardiac remodeling driven by renin overexpression with changes typical for HF such as increased sympathetic tone, LV hypertrophy, myocardial fibrosis and stressrelated pathways. 4,10,12 Results of our study show that the typical course of HF development, characterized by impaired LV relaxation and fast progression (within weeks) to overt HF, was attenuated by galectin-3 inhibition. It remains to be determined if galectin-3 inhibition is equally effective in other multifactorial models of HF such as the spontaneous hypertensive rat or post-myocardial infarction HF, as a single treatment or in addition to established HF therapy. 4 Mechanisms underlying the cardioprotective effects of galectin-3 inhibition Fibrosis is accepted as one of the main determinants of cardiac remodeling. Cessation of the fibrotic process is one of the key targets to reverse cardiac remodeling and improve prognosis. Fibroblasts, together with myofibroblasts and macrophages, have been identified as key cells in the fibrotic process The striking observation that myocardial fibrogenesis was strongly inhibited when galectin-3 was genetically disrupted or pharmacologically inhibited lead us to investigate the effect of galectin-3 on the fibrotic process. First, we showed in REN2 rats that pharmacological inhibition of galectin-3 lead to a lower number of myofibroblasts along with less collagen synthesis (lower PINP plasma levels) and deposition. We substantiated these changes by showing that the stiffness of the fibrotic depositions was also altered. In the collagen digestibility assay, the collagenase digested more collagens in Gal3i-treated REN2 rats (Figure 2e) indicating less cross-linked fibrotic tissue. Studies on for fibrogenesis-related gene profiles in fibroblasts treated with recombinant galectin-3 revealed changes in several genes relevant to the extracellular matrix processing. In vitro incubation of human dermal fibroblasts with galectin-3 resulted in significant upregulation of genes coding 69

71 for various fibrillar collagens (COL1A1, COL1A2 and COL3A1) and genes involved in the modification of (pro)collagens including prolyl-4-hydroxylase (P4HB), heat shock protein 47 (HSP47 or SERPINH1), the procollagen C-endopeptidase enhancer (PCOLCE) and the lysyl oxidases (LOXL2). Inhibition of P4Hs has been shown to inhibit fibrosis and preserve cardiac function in HF Also, the antisense HSP47 was associated with less myocardial fibrosis protecting hearts from post myocardial infarction remodeling. 30 Importantly, treatment with N-acetyllactosamine downregulated the expression of all these genes to baseline levels. These results suggest that galectin-3 may affect several steps involved in fibrogenesis, from enhanced synthesis of procollagen to the regulation of various enzymes involved in the processing of procollagen into mature intra and extracellular collagen. The activation of these pathways is attenuated by galectin-3 inhibition (figure 7). Figure 7 Schematic representation of myocardial fibrogenesis and the effects of galectin-3 and galectin 3 inhibition. Upon cardiac stressors (in our study: AngII, TAC, REN2, and rh-galectin-3 treatment), (cardiac) fibroblasts differentiate into myofibroblasts and collagen production increases. Intracellularly, procollagen chains are processed by various gene products, and then secreted into the interstitium. Here, the procollagens are cleaved to become collagens and further processed, resulting in aggregation, maturation, and cross-linking. Finally, the deposited cross-linked collagens in the myocardium add to myocardial fibrosis and dysfunction. Along this process, galectin-3 exerts effects on genes involved in the various steps of fibrogenesis, and galectin-3 inhibition with N-Lac inhibits these various steps.

72 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling As the effect of galectin-3 on fibroblasts provides an explanation as to why in vivo galectin-3 inhibition results in the cessation of the fibrotic process, we validated these findings in REN2 rats. The cardiac expression of Col1a1, Col1a2, Col3a1, P4hb, Hsp47 (Serpinh1), Pcolce and Loxl2 genes was upregulated and treatment with N-acetyllactosamine normalized their expression to levels similar to those of control SD rats. These observations were further substantiated by showing, both in vitro and in vivo, decreased collagen levels in culture medium (fibroblasts) and plasma (REN2 rats), increased collagen digestibility (figure 2e), regulation of MMPs and TIMPs (Supplemental figure S6) and decreased myocardial fibrosis upon Gal3i treatment (figure 2d). Overall, our results point towards a pivotal role of galectin-3 in cardiac fibrosis. The general effect of galectin-3 on cardiac remodeling, due to myocardial fibrogenesis, appeared to be similar between mice and rats. Furthermore, the same fibrogenic effects of galectin-3 (and its inhibition by N-acetyllactosamine) were observed in primary human fibroblasts and in REN2 rat hearts. Other observations LV weight was increased in Gal3-KO mice and in REN2 rat treated with N- acetyllactosamine. The mechanism(s) underlying this observation has not been elucidated as we were not able to detect appreciable protein expression of galectin-3 in cardiomyocytes in our preliminary experiments. It has also been reported that galectin-3 is mainly produced by macrophages homing to sites of injury. 4,8 We confirmed the influx of macrophages and the colocalization of galectin-3 with macrophages in our REN2 models (Supplemental figures S4 and S5). However, anti-galectin-3 treatment did not reduce the number of macrophages (figure 3d). It has been suggested that the activation of macrophages is more pivotal than the number of macrophages in the development of cardiac remodeling. Usher and colleagues 31 described mice with a deletion of the mineralocorticoid receptor in macrophages and showed they were protected against cardiac remodeling. Our findings do not exclude a role for macrophages in galectin-3-mediated HF but, from the data presented herein, we conclude that Gal3i exerts its effects primarily via binding to the CRD of galectin-3, that prevents a profibrotic effect of activated galectin-3, and not through an increase or decrease in the number of macrophages. 4 Clinical Perspectives Clinical proof for a role of galectin-3 in HF comes from several studies reporting the value of galectin-3 as a biomarker in HF. 6,21,32-34 In the general population, it has been recently observed that sustained elevation in galectin-3 levels may contribute to increased cardiovascular risk, all cause mortality (PREVEND cohort) 35 and new onset HF (Dr. J. Ho, Framingham Heart Study, unpublished data, 2012). Our data lend further support to the role of galectin-3 in cardiac remodeling as well as its potential role as a target for therapy. Future studies are being designed to establish the role of galectin-3 inhibitory compounds in HF of 71

73 Chapter 4 different etiology or on top of established HF therapy. Since high galectin-3 levels may predispose for the development of HF, therapies targeted against galectin-3 may afford protection. Interestingly, although speculative, high intake of dietary cereals rich in pectins that inhibit galectin-3 has been associated with lower risk for new onset HF. 36 Limitations We studied only male animals and, given the established differences in cardiac remodeling between sexes, our results cannot be extrapolated to female animals. Furthermore, results on cardiac contractility and relaxation should be interpreted with caution as no loadindependent measures of LV function were reported (EES, EED, PRSW, etc). Because we only tested one dose of N-acetyllactosamine, dose-finding studies are warranted. Finally, not all fibrotic genes responded to the same extent in our different models. This might be attributed to the differences in etiology, model severity or to differences between mice and rats. Nevertheless, and despite some gene-specific differences, the overall response to galectin-3 interference (inhibition and knock-out) was similar. Conclusions Genetic disruption of galectin-3 and pharmacological inhibition of galectin-3 attenuated the progression of cardiac remodeling in murine and rat models of HF. Inhibition of galectin-3 largely preserved systolic and diastolic function via the inhibition of myocardial fibrosis and decreased collagen production, processing and cross-linking. Future, more in depth, mechanistic studies would be needed to address the precise role of galectin-3 in HF development. At later stages of remodeling, galectin-3 inhibition prevented further HF progression. Taken together, our results strongly suggest a causal role of galectin-3 in the development of cardiac remodeling and HF and we postulate that galectin-3-targeted therapy may potentially be a useful addendum in the treatment of HF. Acknowledgment We thank Danielle Libersan, PhD for her assistance in preparing this manuscript. We thank Inge Vreeswijk-Baudoin and Martin Dokter for excellent technical assistance. Funding Sources Parts of these studies were funded by BG Medicine Inc. (Waltham, MA, USA) who provided an unrestricted research grant to the department of Cardiology of the University Medical Center Groningen, The Netherlands. Dr. de Boer is supported by the Innovational Research Incentives Scheme program of the Netherlands Organization for Scientific Research (NWO VENI, grant ) and the Netherlands Heart Foundation (Grant 2007T046). Disclosures 72

74 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling BG Medicine Inc. owns certain rights with respect to the use of galectin-3 as a biomarker. Dr. de Boer and Dr. van Veldhuisen received honoraria from BG Medicine. BG Medicine provided research grants to the department of Cardiology of the University Medical Center Groningen, the Netherlands. 4 73

75 Chapter 4 References 1. Yang RY, Rabinovich GA, Liu FT. Galectins: structure, function and therapeutic potential. Expert Rev Mol Med. 2008; 10: e Henderson NC, Mackinnon AC, Farnworth SL, Kipari T, Haslett C, Iredale JP, Liu FT, Hughes J, Sethi T. Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol. 2008; 172: Kolatsi-Joannou M, Price KL, Winyard PJ, Long DA. Modified citrus pectin reduces galectin-3 expression and disease severity in experimental acute kidney injury. PLoS One. 2011; 6: e Sharma UC, Pokharel S, van Brakel TJ, van Berlo JH, Cleutjens JP, Schroen B, Andre S, Crijns HJ, Gabius HJ, Maessen J, Pinto YM. Galectin-3 marks activated macrophages in failure-prone hypertrophied hearts and contributes to cardiac dysfunction. Circulation. 2004; 110: Liu YH, D'Ambrosio M, Liao TD, Peng H, Rhaleb NE, Sharma U, Andre S, Gabius HJ, Carretero OA. N-acetyl-serylaspartyl-lysyl-proline prevents cardiac remodeling and dysfunction induced by galectin-3, a mammalian adhesion/growthregulatory lectin. Am J Physiol Heart Circ Physiol. 2009; 296: H de Boer RA, Voors AA, Muntendam P, van Gilst WH, van Veldhuisen DJ. Galectin-3: a novel mediator of heart failure development and progression. Eur J Heart Fail. 2009; 11: de Couto G, Ouzounian M, Liu PP. Early detection of myocardial dysfunction and heart failure. Nat Rev Cardiol. 2010; 7: Henderson NC, Mackinnon AC, Farnworth SL, Poirier F, Russo FP, Iredale JP, Haslett C, Simpson KJ, Sethi T. Galectin- 3 regulates myofibroblast activation and hepatic fibrosis. Proc Natl Acad Sci U S A. 2006; 103: Hsu DK, Yang RY, Pan Z, Yu L, Salomon DR, Fung-Leung WP, Liu FT. Targeted disruption of the galectin-3 gene results in attenuated peritoneal inflammatory responses. Am J Pathol. 2000; 156: de Boer RA, Pokharel S, Flesch M, van Kampen DA, Suurmeijer AJ, Boomsma F, van Gilst WH, van Veldhuisen DJ, Pinto YM. Extracellular signal regulated kinase and SMAD signaling both mediate the angiotensin II driven progression towards overt heart failure in homozygous TGR(mRen2)27. J Mol Med (Berl). 2004; 82: Ruifrok WP, Qian C, Sillje HH, van Goor H, van Veldhuisen DJ, van Gilst WH, de Boer RA. Heart failure-associated anemia: bone marrow dysfunction and response to erythropoietin. J Mol Med (Berl). 2011; 89: Lee MA, Bohm M, Paul M, Bader M, Ganten U, Ganten D. Physiological characterization of the hypertensive transgenic rat TGR(mREN2)27. Am J Physiol. 1996; 270: E Rockman HA, Wachhorst SP, Mao L, Ross J,Jr. ANG II receptor blockade prevents ventricular hypertrophy and ANF gene expression with pressure overload in mice. Am J Physiol. 1994; 266: H Demotte N, Wieers G, Van Der Smissen P, Moser M, Schmidt C, Thielemans K, Squifflet JL, Weynand B, Carrasco J, Lurquin C, Courtoy PJ, van der Bruggen P. A galectin-3 ligand corrects the impaired function of human CD4 and CD8 tumor-infiltrating lymphocytes and favors tumor rejection in mice. Cancer Res. 2010; 70: Kuipers I, Li J, Vreeswijk-Baudoin I, Koster J, van der Harst P, Sillje HH, Kuipers F, van Veldhuisen DJ, van Gilst WH, de Boer RA. Activation of liver X receptors with T attenuates cardiac hypertrophy in vivo. Eur J Heart Fail. 2010; 12: Pacher P, Nagayama T, Mukhopadhyay P, Batkai S, Kass DA. Measurement of cardiac function using pressure-volume conductance catheter technique in mice and rats. Nat Protoc. 2008; 3:

76 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling 17. Creemers LB, Jansen DC, van Veen-Reurings A, van den Bos T, Everts V. Microassay for the assessment of low levels of hydroxyproline. BioTechniques. 1997; 22: Kuipers I, van der Harst P, Kuipers F, van Genne L, Goris M, Lehtonen JY, van Veldhuisen DJ, van Gilst WH, de Boer RA. Activation of liver X receptor-alpha reduces activation of the renal and cardiac renin-angiotensin-aldosterone system. Lab Invest. 2010; 90: Badenhorst D, Maseko M, Tsotetsi OJ, Naidoo A, Brooksbank R, Norton GR, Woodiwiss AJ. Cross-linking influences the impact of quantitative changes in myocardial collagen on cardiac stiffness and remodelling in hypertension in rats. Cardiovasc Res. 2003; 57: van der Slot-Verhoeven AJ, van Dura EA, Attema J, Blauw B, Degroot J, Huizinga TW, Zuurmond AM, Bank RA. The type of collagen cross-link determines the reversibility of experimental skin fibrosis. Biochim Biophys Acta. 2005; 1740: de Boer RA, Lok DJ, Jaarsma T, van der Meer P, Voors AA, Hillege HL, van Veldhuisen DJ. Predictive value of plasma galectin-3 levels in heart failure with reduced and preserved ejection fraction. Ann Med. 2011; 43: Sorme P, Qian Y, Nyholm PG, Leffler H, Nilsson UJ. Low micromolar inhibitors of galectin-3 based on 3'-derivatization of N-acetyllactosamine. Chembiochem. 2002; 3: Umemoto K, Leffler H, Venot A, Valafar H, Prestegard JH. Conformational differences in liganded and unliganded states of Galectin-3. Biochemistry. 2003; 42: Berk BC, Fujiwara K, Lehoux S. ECM remodeling in hypertensive heart disease. J Clin Invest. 2007; 117: Kakkar R, Lee RT. Intramyocardial fibroblast myocyte communication. Circ Res. 2010; 106: Souders CA, Bowers SL, Baudino TA. Cardiac fibroblast: the renaissance cell. Circ Res. 2009; 105: Myllyharju J. Prolyl 4-hydroxylases, key enzymes in the synthesis of collagens and regulation of the response to hypoxia, and their roles as treatment targets. Ann Med. 2008; 40: Fielitz J, Philipp S, Herda LR, Schuch E, Pilz B, Schubert C, Gunzler V, Willenbrock R, Regitz-Zagrosek V. Inhibition of prolyl 4-hydroxylase prevents left ventricular remodelling in rats with thoracic aortic banding. Eur J Heart Fail. 2007; 9: Nwogu JI, Geenen D, Bean M, Brenner MC, Huang X, Buttrick PM. Inhibition of collagen synthesis with prolyl 4- hydroxylase inhibitor improves left ventricular function and alters the pattern of left ventricular dilatation after myocardial infarction. Circulation. 2001; 104: Hagiwara S, Iwasaka H, Shingu C, Matumoto S, Hasegawa A, Noguchi T. Heat shock protein 47 (HSP47) antisense oligonucleotides reduce cardiac remodeling and improve cardiac function in a rat model of myocardial infarction. Thorac Cardiovasc Surg. 2011; 59: Usher MG, Duan SZ, Ivaschenko CY, Frieler RA, Berger S, Schutz G, Lumeng CN, Mortensen RM. Myeloid mineralocorticoid receptor controls macrophage polarization and cardiovascular hypertrophy and remodeling in mice. J Clin Invest. 2010; 120: van Kimmenade RR, Januzzi JL,Jr, Ellinor PT, Sharma UC, Bakker JA, Low AF, Martinez A, Crijns HJ, MacRae CA, Menheere PP, Pinto YM. Utility of amino-terminal pro-brain natriuretic peptide, galectin-3, and apelin for the evaluation of patients with acute heart failure. J Am Coll Cardiol. 2006; 48:

77 Chapter Lok DJ, Van Der Meer P, de la Porte PW, Lipsic E, Van Wijngaarden J, Hillege HL, van Veldhuisen DJ. Prognostic value of galectin-3, a novel marker of fibrosis, in patients with chronic heart failure: data from the DEAL-HF study. Clin Res Cardiol. 2010; 99: Shah RV, Chen-Tournoux AA, Picard MH, van Kimmenade RR, Januzzi JL. Galectin-3, cardiac structure and function, and long-term mortality in patients with acutely decompensated heart failure. Eur J Heart Fail. 2010; 12: de Boer RA, van Veldhuisen DJ, Gansevoort RT, Muller Kobold AC, van Gilst WH, Hillege HL, Bakker SJ, van der Harst P. The fibrosis marker galectin-3 and outcome in the general population. J Intern Med. 2012; 272: Djousse L, Gaziano JM. Breakfast cereals and risk of heart failure in the physicians' health study I. Arch Intern Med. 2007; 167:

78 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Supplemental methods Animal models We studied six- to ten-week old male mice deficient for the gene encoding galectin-3 (galectin-3 knock-out mice (Gal3-KO)). The Gal3-KO homozygous mice carrying a targeted mutation in galectin-3 (a lectin, galactose binding soluble 3 or Lgals3) have been generated and are bred at Jackson Laboratory (Bar Harbor, ME, USA) and shipped to Groningen, The Netherlands. A targeting vector containing neomycin resistance and herpes simplex virus thymidine kinase genes was used to disrupt 3.7kb of sequence that includes exons 2, 3 and 4. The construct was electroporated into WW6 embryonic stem (ES) cells (derived from 129/Sv, C57BL/6 and SJL mixed background mice). Correctly targeted ES cells were injected into outbred MF-1 blastocysts. The resulting chimeric male animals were crossed with 129 female mice and the mutant strain was backcrossed with C57BL/6 for five generations. For an overview of the generation of the Gal3-KO mice, see Supplemental Figure S1. Other researchers have generated Gal3-KO mice as well. 1-3 Transgene-negative male wild-type (WT) served as controls. For the reversal experiment, we used six- to eight-week old male C56Bl6/J mice (Harlan, The Netherlands). We also studied six-week old, male, homozygous TGR(mREN2)27 rats (REN2). These rats overexpress the mouse renin-2 gene (ren-2d) and have a phenotype of severe hypertension and left ventricular (LV) hypertrophy which culminates into heart failure (HF) and low blood pressure over the course of weeks. 4-7 The rats were bred at the Max Delbrück Center for Molecular Medicine (Berlin Buch, Germany). Since Sprague Dawley (SD) rats represent the appropriate control strain for REN2 rats, age-matched male SD rats were used as controls (Harlan, The Netherlands). 6 Animals were housed under standard condition. All animal studies were approved by the Animal Ethical Committee of the University of Groningen, The Netherlands, and conducted in accordance with existing guidelines for the care and use of laboratory animals. 4 Prevention study - mouse models We induced cardiac remodeling by two interventions. First, cardiac remodeling was provoked by subcutaneous administration of angiotensin II (AngII) via osmotic minipumps for 14 days. Second, we induced pressure overload and cardiac remodeling by transverse aortic constriction (TAC) for 28 days. The control group received saline infusion via osmotic minipumps for 14 days. In total, six groups were studied: WT-con (control, N=10), Gal3-KOcon (N=10), WT-AngII (N=12), Gal3-KO-AngII (N=10), WT-TAC (N=12) and Gal3-KO- TAC (N=14). AngII (Bachem AG, Bubendorf, Switzerland) was dissolved in 0.9% NaCl and injected into osmotic minipumps (Alzet 2004, Durect Corporation, Cupertino, CA, USA). The dose delivered by the osmotic minipumps was 2.5 µg/kg/day. The minipumps were inserted subcutaneously on the back of the mice. 77

79 Chapter 4 The TAC is a well established model. 8 In brief, mice were anesthetized with oxygen and isoflurane (2%), intubated and ventilated. The thoracic cavity was opened between the second and the third intercostal cavity. Then, a blunted needle (27G) was placed on the aortic arch between both carotid arteries and, with an 8-0 nylon suture, the aorta was tied onto the needle. Immediately after, the needle was removed creating a reproducible stenosis of the aorta of about 50%. Sham-operated mice served as controls. Mice were sacrificed after 28 days. Reversal study - mouse model The TAC model was used as described above. Mice were left untreated for 28 days. After 28 days, one third of all animals were sacrificed. The remainder were given intraperitoneal injections of the galectin-3 inhibitor (Gal3i) N-acetyllactosamine (N-Lac) (Sigma-Aldrich, Zwijndrecht, The Netherlands) that targets the carbohydrate recognition domain (CRD) of galectin-3 at a final dose of 5 mg/kg/day three times per week. N-Lac has been shown to effectively bind galectin-3. 9 The treatment period lasted another 28 days and the total study duration was 56 days (eight weeks). Control mice were injected with saline. In total, three groups were studied: WT-TAC-control (N=9), WT-TAC-saline (N=9) and WT-TAC-Gal3i (N=9). Rat model To study the effect of galectin-3 binding protein on cardiac remodeling, we allocated SD and REN2 rats to different treatment regimens. REN2 rats were treated with N-Lac at a final dose of 5 mg/kg/day injected intraperitoneally three times per week. SD rats were used as controls for all groups. Rats were sacrificed at week seven of the experimental protocol (when aged ~13 weeks). In total, four groups were studied: SD-con (control, N=12), REN2-con (N=18), SD-Gal3i (N=8) and REN2-Gal3i (N=10). Treatment with Gal3i did not exert any effects in control SD rats and, therefore, the results shown are limited to the REN2 groups. Echocardiography Cardiac function was assessed by echocardiography at baseline and prior to sacrifice with Vivid 7 (GE Healthcare, Chalfont St Giles, UK) equipped with a 10-MHz (rats) and a 13- MHz (mice) phase array linear transducer), as described previously. 10 The echocardiographic measurements were performed under general anesthesia with 2% isoflurane. Both 2- dimensional (2D) images in parasternal long-axis and short-axis view and 2-D guided M- mode tracing were obtained. Parasternal long-axis (PLAX) views were obtained in order to ensure that the mitral and aortic valves and the apex were visualized. Left ventricular outflow tract (LVOT) diameter was measured in PLAX. Short-axis views were recorded at the level of mid-papillary muscles. A total of three loops was recorded and used for calculations. LV endsystolic diameter (LVESD) and LV end-diastolic diameter (LVEDD) were measured from M- mode tracings. LV fractional shortening was calculated as (LVEDD LVESD)/LVEDD 100%. Cardiac output was calculated by echo Doppler measurements over the aortic valve 78

80 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling (pulsed wave doppler) using the LVOT diameter. To monitor the development of cardiac hypertrophy in both mice and rats, echocardiography was performed at baseline and prior to sacrifice. Measurement of hemodynamic function Hemodynamic function was assessed invasively as described previously 11 by introducing a 0.8-French (for mice) and a 1.4 French (for rats) microtip pressure-volume transducer (Millar Instr. Inc., Houston, TX, USA) via the right carotid artery into the aorta. A three minute period was allowed for stabilization and then systolic and diastolic blood pressure and heart rate (HR) were recorded in the aorta (average of 20 heart cycles). The catheter was then advanced into the LV. Peak systolic pressure, LV end diastolic pressure (LVEDP), maximal indices of LV contraction and relaxation (dpdtmax and dpdtmin) as well as the relaxation constant Tau were measured. dpdtmax and dpdtmin were corrected for peak systolic pressure to correct for differences in afterload, as described previously. 4 Tissue procurement After measuring hemodynamics, blood was drawn (PINP assay) and the hearts were rapidly excised and weighed. Myocardial tissue was dissected transversally and processed for immunohistochemistry or snap frozen for molecular analysis. Immunohistochemistry Hearts were isolated and fixed with buffered 3.7% formalin for 24 hours. Thereafter, tissue sections were dehydrated and embedded in paraffin. Another slice was embedded in tissue-tec, placed on a carton disk put on a petri-dish and immediately, but slowly, frozen above liquid nitrogen. For immunohistochemistry, 3 µm-thick sections were cut on a microtome and mounted on slides. Hearts were cut at the mid-papillary level. To measure fibrosis, Masson s trichrome staining was performed on paraffin sections for all experimental animals. Whole stained sections were scanned (Nanozoomer 2.0-HT, Hamamatsu, Japan) and fibrosis score for the entire section was calculated at 20x magnification for mice and 10x magnification for rats (ScanScope, Aperio Technologies, Vista, CA, USA). To measure cardiomyocyte size, deparaffinized sections were stained with Gomori's silver staining. 12 Cardiomyocyte size was measured as the cross-sectional area of transversally cut cardiomyocytes at 40x magnification for rats. Myofibroblasts were stained with a monoclonal mouse anti-human smooth muscle actin (SMA) antibody (#M0851, DAKO, Heverlee, Belgium). SMA in the entire section was calculated at 10x magnification by the SanScope software for rats. To measure the number of macrophages, macrophages were stained with a mouse anti-rat CD68 (ED1) antibody (#MCA341R, AbD Serotec, Oxford, UK) for rats. The number of macrophages was calculated for the entire section using the SanScope software at 40x magnification for rats. Proliferating cells were stained with a mouse monoclonal anti-pcna antibody [PC10]-proliferation marker (#ab29, Abcam, Cambridge, 4 79

81 Chapter 4 UK). The number of proliferating cells in the entire section was calculated by the SanScope software at 40x magnification for rats. To stain galectin-3 in cardiac tissue, an anti-galectin-3 monoclonal antibody (Thermo Fisher Scientific, Landsmeer, The Netherlands) was used. Briefly, paraffin sections were dewaxed and subjected to an antigen retrieval procedure by incubating overnight at 80 C in 0.1M Tris/HCl, ph 9.0. After three washes with TBS, endogenous peroxidase was blocked and sections were incubated with the primary antibody (anti-galectin-3 antibody, 1:100) in 1% BSA/PBS for 1 hour at room temperature. The sections were then incubated with a secondary antibody (polyclonal rabbit anti-mouse IgG/HRP, 1:100) diluted in 1% BSA/PBS buffer for two hours. The slides were first stained with 3-animo-9-ethylcarbazole and counterstained with hematoxillin. For typical examples of galectin-3 staining in cardiac tissue, see supplemental results (Supplemental Figures S3 and S4). Enzyme-linked immunosorbent assay (ELISA) Procollagen type I N-terminal propeptide (PINP), a marker of collagen metabolism, was analyzed in plasma using a commercial Enzyme Linked Immunosorbent Assay according to the manufacturer s instructions (rat procollagen I N-terminal propeptide ELISA kit #E90957Ra, Uscn Life Science Inc, Wuhan, China). Human collagen type III (Human Collagen Type III (Col III) ELISA Kit, Cat No. Hu9614, TSZ ELISA, Framingham, USA) was analyzed in human dermal fibroblast supernatant (medium). Collagen cross-linking assay Collagen digestibility was measured following the digestion of tissue samples (for 6 hours at 37 C) with Clostridium histolyticum collagenase (Sigma) at a final concentration of 5 µg enzyme/mg tissue in a 50 mm Tris buffer containing 5 mm CaCl2, 0.15 M NaCl, 1 µm ZnCl2, 0.02% (w/v) NaN3 and 0.01% (v/v) Brij 35. After incubation, the supernatant and remaining tissue were separated and hydrolyzed in 6 M HCl at 110 C for hours. The relative amount of collagen in the supernatant and remaining tissue was estimated by measuring the amount of hydroxyproline in the hydrolysates of both the supernatant and the remaining tissue. 13 Quantitative real-time PCR cdna synthesis was performed with 0.5 µg total RNA using a specific cdna synthesis kit according to the manufacturer s protocol (Quantitect Rev. Transcriptase kit, Qiagen, Venlo, The Netherlands), as described previously. 10,14,15 Quantitative real-time PCR (RT-qPCR) was performed using SYBR Green mix according to the manufacturer s protocol (Absolute SYBR Green ROX mix, Thermo Scientific, Breda, The Netherlands) on C1000 Thermal Cycler CFX384 Real-Time PCR Detection System (Bio-Rad Laboratories, Veenendaal, The Netherlands). All targets were evaluated under the same experimental conditions (95 C for 15 minutes, then 36 cycles at 95 C for 15 seconds and 60 C for 30 seconds). Samples were 80

82 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling analyzed by quantification software (Bio-Rad CFX Manager 1.6). mrna levels were expressed in relative units based on a standard curve obtained with serial dilutions of a calibrator cdna mixture. To normalize expression data, reference genes were used. Reference genes were chosen with little sample-to-sample variability (GAPDH and 36B4). See Supplemental Tables S1, S2, and S3 for a list of primers used. Western blotting Rat hearts were homogenized in ice-cold RIPA (50 mm Tris ph 8.0, 1% nonidet P40, 0.5% deoxycholate, 0.1% SDS, 150 mm NaCl) containing phosphatase inhibitor cocktail 1 (Sigma), protease inhibitor (ROCHE), 1 mm phenylmethylsulfonyl fluoride (PMSF) and 15 mm NaVanadate for 30 min. Equal amounts of protein (20 µg) were loaded on 12% polyacrylamide gels. After electrophoresis, the gels were blotted onto nitrocellulose membranes. Membranes were then incubated with anti-galectin-3 antibody (Thermo Fisher Scientific, Landsmeer, The Netherlands), anti-hsp47 antibody (#ab109117, Abcam, Cambridge, UK), anti-mac-2 antibody (CL8942AP, Cedarlanelabs, USA), troponin T (T6277, Sigma, Netherlands), anti-vimentin antibody (V5255, Sigma, Netherlands), anti-cd68 antibody (ab76308, Abcam, Cambridge, UK) and anti-gapdh antibody (10R-G109a, Fitzgerald, USA) overnight at 4 C. Membranes were washed and incubated with the appropriate secondary HRP-conjugated antibody and signals were visualized with ECL and analyzed with densitometry (Syngene, Cambridge, United Kingdom). 4,14,15 4 Human adult dermal fibroblast culture experiments Human adult dermal fibroblasts (primary cells) were cultured for 72 hours in culture medium (EMEM containing 1% Pen/Strep, 1% L-glutamine, 0.5% FCS and mm ascorbid acid 2-phosphate) with or without galectin-3 (10 µg/ml; #450-38, Preprotech, Rocky Hill, NY, USA), N-Lac (10 mm) or a combination. The initial seeding density was 15,000 cells/cm 2. Cells were harvested and RNA was isolated. cdna synthesis was conducted using RevertAid First Strand cdna Synthesis Kit (#K1622, Fermentas, St. Leon-Rot, Germany) and cdna was hybridized to a custom made low density array carrying probes for ~50 genes involved in fibrosis (collagen production, collagen cross-linking, collagen processing and collagen degradation). The complete list of gene is presented in Table S4 of the Supplemental data. Genes that were most significantly expressed (LOXL2, HSP47, PCOLCE, COL1A1, COL1A2, COL3A1, and P4HB) were verified using quantitative RT-qPCR and normalized against the reference gene GAPDH. In addition, vimentin staining, a fibroblast marker, was carried out with anti-vimentin (C-20; Santa Cruz Biotechnology, Santa Cruz, CA, USA) and protein levels of collagen type 3 were measured with ELISA (human collagen 3 ELISA kit, #Hu9614, TSZ Scientific LLC, Framingham, MA, USA). Collagen type 3 protein levels were measured in singulo, therefore error bars and statistics are not provided. 81

83 Chapter 4 Analyzes of galectin-3 expression in macrophages and in adult rat cardiac myocytes and fibroblast Adult rat cardiomyocytes and fibroblasts were isolated using a Langendorff perfusion system with retrograde perfusion. 16 The perfusion solution (10 mm HEPES (ph 7.4), mm NaCl, 4.0 mm KCl, 1.2 mm NaH 2 PO 4, 1.2 mm MgSO 4, 20 µm CaCl 2 ) contained 0.2mg/ml Liberase TM Research Grade (Roche) to digest the extracellular matrix. Cardiomyocytes were subsequently separated from other cells by several rounds of velocity sedimentation in a perfusion solution containing 1mg BSA/ml. Adult cardiomyocytes were directly frozen after isolation. Cardiac fibroblasts were cultured for several days in DMEM medium supplemented with 10% FCS and penicillin-streptomycin (100 IU/ml and 100ug/ml, respectively) to obtain sufficient cells for analysis. A rat alveolar macrophage cell line 17 was also cultured in the same medium. Cell lysates were generated as described in the Western blot section and equal amounts of proteins were subsequently loaded on a SDS-PAGE gel. 82

84 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Supplemental tables Table S1. List of primers used for RT-qPCR in mice RT-qPCR primer, 5 to 3 Gene symbol (name) Forward Reverse Nppa (ANP) ATGGGCTCCTTCTCCATCAC TCTACCGGCATCTTCTCCTC Lgals3 (Galectin-3) TATCCTGCTGCTGGCCCTTATG GTTTGCGTTGGGTTTCACTG Rplp0 (36B4) AAGCGCGTCCTGGCATTGTC GCAGCCGCAAATGCAGATGG Table S2. List of primers used for RT-qPCR in rats RT-qPCR primer, 5 to 3 Gene symbol (name) Forward Reverse Nppa (ANP) ATGGGCTCCTTCTCCATCAC TCTACCGGCATCTTCTCCTC Lgals3 (Galectin-3) CCCGCTTCAATGAGAACAAC ACCGCAACCTTGAAGTGGTC Loxl2 TGACTGCCAGTGGATAGAC ATGCGGTAGCCATCATAGC Hsp47 TCATGGTGACCCGCTCCTAC GCTTATGGGCCAAGGGCATC Pcolce GAAGAAAGGAGCCAGTTACC GGGCACTTTCTCTTGCTTAG Col1a1 ACAGCGTAGCCTACATGG AAGTTCCGGTGTGACTCG Col1a2 ATGGTGGCAGCCAGTTTG GCTGTTCTTGCAGTGGTAGG Col3a1 TGGAAACCGGAGAAACATGC CAGGATTGCCATAGCTGAAC P4hb ACCAGCGCATACTTGAGTTC ACTCCGGTTTGTACTTGGTC Gapdh CATCAAGAAGGTGGTGAAGC ACCACCCTGTTGCTGTAG Mmp2 TGAGCTCCCGGAAAAGATTG CATTCCCTGCGAAGAACACA Mmp9 CGGGAACGTATCTGGAAATTCG CATGGCAGAAATAGGCCTTGTC Timp1 AGAGCCTCTGTGGATATGTC CTCAGATTATGCCAGGGAAC Timp2 TGGACGTTGGAGGAAAGAAG TGTCCCAGGGCACAATAAAG 4 83

85 Chapter 4 Table S3. List of primers used for RT-qPCR in human cell culture experiments RT-qPCR primer, 5 to 3 Gene symbol (name) Forward Reverse LOXL2 CCGCATCTGGATGTACAAC TTAAGAGCCCGCTGAAGTG HSP47 CTTCATGGTGACTCGGTCCT CGATTTGCAGCTTTTCCTTC PCOLCE CTGTTCCTGGCACATCATC CTTGTAGGAGGCTGAGAAG COL1A1 GCCTCAAGGTATTGCTGGAC ACCTTGTTTGCCAGGTTCAC COL1A2 CTGGAGAGGCTGGTACTGCT AGCACCAAGAAGACCCTGAG COL3A1 CTGGACCCCAGGGTCTTC CATCTGATCCAGGGTTTCCA P4HB TTGATGAAGGCCGGAACAAC TCTGCTCGGTGAACTCGATG GAPDH CATCAAGAAGGTGGTGAAGC ACCACCCTGTTGCTGTAG 84

86 Table S4. List of genes of which probes are carried by the Low Density Array Verified by RT-qPCR Upregulated by Galectin-3 and downregulated by N-LAC Downregulated by N-Lac Number Gene Upregulated by Galectin-3 Collagen modifying enzymes 1 PLOD PLOD PLOD P4HA P4HA2 + + * 6 P4HA P4HB + + * ** 8 LEPRE1 + + * 9 LEPREL1 + + * 10 LEPREL LOX LOXL LOXL2 + + * ** 14 LOXL SERPHINH1 + + * ** 16 ADAMTS2 + + * 17 ADAMTS3 + + * 18 ADAMTS * 19 BMP MEP1A MEP1B + -

87 22 PCOLCE + + * ** 23 PCOLCE SNAI GLT25D GLT25D PPIB FKBP * 29 SLC39A CRTAP - - Collagens 31 COL1A1 + + * ** 32 COL1A2 + + * ** 33 COL2A COL3A1 + + * ** 35 COL4A1 + + * 36 COL5A1 + + * Proteinases 37 MMP MMP MMP MMP * 41 TIMP CTSK + - Collagen receptors 43 DDR1 + + * 44 DDR Endo 180 (MRC2) - + +: at least 25% change by any perturbations.

88 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Supplemental figures and figure legends Figure S1. Strategy for the genetic disruption of galectin-3 in mice and analysis of galectin-3 expression. 4 Figure S1 Shows schedule of creation of galectin-3 knock-out (Gal3-KO) mice and expression of galectin-3 in the different mouse groups. (A) Exon 2, 3 and 4 were disrupted in the Gal3-KO mice. (B) mrna expression of galectin-3 in the left ventricles of mice from the different experimental groups (n= 7-12). (C) Representative Western blot showing galectin-3 expression (anti-mac-2) in left ventricles of mice from the different experimental groups. (D) Quantification of Western blot (n=5 ). Con: control (saline-infused); AngII: angiotensin II; sham: sham-operated mice; TAC: transverse aortic constriction. * P<0.05 vs. WT-con (BON, M=6 tests), P<0.05 vs. WT-AngII (BON, M=6 tests), P<0.05 vs. WT-sham (BON, M=6 tests), ** P<0.05 vs. WT-TAC (BON, M=6 tests). 87

89 Chapter 4 Figure S2. Survival of REN2 rats in the different groups. Figure S2 shows the survival of REN2 rats in the different experimental groups. In the REN2-con group, 7 out of 18 rats (39%) died and 1 out of 10 rats (10%)died in the REN2-Gal3i group. No SD control animals died (SDcon group). Con: control; Gal3i: galectin-3 inhibitor. Figure S3. Localization of galectin-3 expression in cardiac tissue. Figures S3 shows the localization of galectin-3 in cardiac tissue from REN2 and SD rats. Immunohistochemical analysis showed that galectin-3 immunoreactivity (red-brown color) is predominantly observed at interstitial sites but not in cardiomyocytes. Con: control; Gal3i: galectin-3 inhibitor. 88

90 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Figure S4. Localization of galectin-3. Panels A-C: Lower magnification (20X) showing double labeling for galectin-3 (red), the macrophage marker ED1 (green) and the merged image with DAPI staining nuclei (blue) in the different experimental groups. Panel D: higher magnification (40X) of the merged image of galectin-3, the macrophage marker ED1 and DAPI (blue). Panels E-G: Lower magnification (20X) showing double labeling for galectin-3 (red), the fibroblast marker collagen III (green) and the merged image with DAPI staining nuclei (blue) in the different groups. Panel H: higher magnification (40X) of the merged image of galectin-3, the fibroblast marker collagen III and DAPI (blue). Con: control; Gal3i: galectin-3 inhibitor. 4 Figure S5. Protein expression of galectin-3 in cardiomyocytes, fibroblasts and macrophages. Figure S5 shows a Western blot with lysates from different cell types. Galectin-3 was detected using antigalectin-3, anti-troponin T (cardiomyocyte marker), anti-vimentin (fibroblast marker) and anti-cd68 (macrophage marker) antibodies and an anti-gapdh antibody (as loading control). CMC = adult rat cardiomyocyte; FB = fibroblast and macrophage = a rat alveolar macrophage cell line. 89

91 Chapter 4 Figure S6. Extracellular matrix protein gene expression in rat hearts. Figure S6 shows changes in gene expression of extracellular matrix proteins in rat hearts as determined by RTqPCR. (A) Timp1, (B) Timp2, (c) Mmp2 and (d) Mmp9. Gal3i: galectin-3 inhibitor. P<0.05 SD-con vs. REN2-con (BON, M=3 tests), P<0.05 SD-con vs. REN2-Gal3i (BON, M=3 tests). 90

92 Genetic and Pharmacological Inhibition of Galectin-3 Prevents Cardiac Remodeling Supplemental References 1. Henderson NC, Mackinnon AC, Farnworth SL, Poirier F, Russo FP, Iredale JP, Haslett C, Simpson KJ, Sethi T. Galectin-3 regulates myofibroblast activation and hepatic fibrosis. Proc Natl Acad Sci U S A. 2006; 103: Henderson NC, Mackinnon AC, Farnworth SL, Kipari T, Haslett C, Iredale JP, Liu FT, Hughes J, Sethi T. Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol. 2008; 172: Hsu DK, Yang RY, Pan Z, Yu L, Salomon DR, Fung-Leung WP, Liu FT. Targeted disruption of the galectin-3 gene results in attenuated peritoneal inflammatory responses. Am J Pathol. 2000; 156: de Boer RA, Pokharel S, Flesch M, van Kampen DA, Suurmeijer AJ, Boomsma F, van Gilst WH, van Veldhuisen DJ, Pinto YM. Extracellular signal regulated kinase and SMAD signaling both mediate the angiotensin II driven progression towards overt heart failure in homozygous TGR(mRen2)27. J Mol Med (Berl). 2004; 82: de Borst MH, Navis G, de Boer RA, Huitema S, Vis LM, van Gilst WH, van Goor H. Specific MAP-kinase blockade protects against renal damage in homozygous TGR(mRen2)27 rats. Lab Invest. 2003; 83: Lee MA, Bohm M, Paul M, Bader M, Ganten U, Ganten D. Physiological characterization of the hypertensive transgenic rat TGR(mREN2)27. Am J Physiol. 1996; 270: E Ruifrok WP, Qian C, Sillje HH, van Goor H, van Veldhuisen DJ, van Gilst WH, de Boer RA. Heart failureassociated anemia: bone marrow dysfunction and response to erythropoietin. J Mol Med (Berl). 2011; 89: Rockman HA, Wachhorst SP, Mao L, Ross J,Jr. ANG II receptor blockade prevents ventricular hypertrophy and ANF gene expression with pressure overload in mice. Am J Physiol. 1994; 266: H Demotte N, Wieers G, Van Der Smissen P, Moser M, Schmidt C, Thielemans K, Squifflet JL, Weynand B, Carrasco J, Lurquin C, Courtoy PJ, van der Bruggen P. A galectin-3 ligand corrects the impaired function of human CD4 and CD8 tumor-infiltrating lymphocytes and favors tumor rejection in mice. Cancer Res. 2010; 70: Kuipers I, Li J, Vreeswijk-Baudoin I, Koster J, van der Harst P, Sillje HH, Kuipers F, van Veldhuisen DJ, van Gilst WH, de Boer RA. Activation of liver X receptors with T attenuates cardiac hypertrophy in vivo. Eur J Heart Fail. 2010; 12: Pacher P, Nagayama T, Mukhopadhyay P, Batkai S, Kass DA. Measurement of cardiac function using pressure-volume conductance catheter technique in mice and rats. Nat Protoc. 2008; 3: van der Meer P, Lipsic E, Henning RH, Boddeus K, van der Velden J, Voors AA, van Veldhuisen DJ, van Gilst WH, Schoemaker RG. Erythropoietin induces neovascularization and improves cardiac function in rats with heart failure after myocardial infarction. J Am Coll Cardiol. 2005; 46: Creemers LB, Jansen DC, van Veen-Reurings A, van den Bos T, Everts V. Microassay for the assessment of low levels of hydroxyproline. BioTechniques. 1997; 22:

93 Chapter Kuipers I, van der Harst P, Kuipers F, van Genne L, Goris M, Lehtonen JY, van Veldhuisen DJ, van Gilst WH, de Boer RA. Activation of liver X receptor-alpha reduces activation of the renal and cardiac reninangiotensin-aldosterone system. Lab Invest. 2010; 90: Lu B, Tigchelaar W, Ruifrok WP, van Gilst WH, de Boer RA, Sillje HH. DHRS7c, a novel cardiomyocyteexpressed gene that is down-regulated by adrenergic stimulation and in heart failure. Eur J Heart Fail. 2012; 14: Louch WE, Sheehan KA, Wolska BM. Methods in cardiomyocyte isolation, culture, and gene transfer. J Mol Cell Cardiol. 2011; 51: Lane KB, Egan B, Vick S, Abdolrasulnia R, Shepherd VL. Characterization of a rat alveolar macrophage cell line that expresses a functional mannose receptor. J Leukoc Biol. 1998; 64:

94

95

96 Chapter 5 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Lili Yu; Anne-Roos S. Frenay; A. Rogier van der Velde; I. Baudoin Vreeswijk; Wiek H. van Gilst; Harry van Goor; Herman H.W. Silljé; Rudolf A. de Boer Manuscript

97 Chapter 5 Abstract Background: Galectin-3 activation has been implicated in renal damage and fibrogenesis. Limited data are available to suggest that galectin-3 targeted intervention acts as a potential therapeutic candidate for the prevention of chronic kidney disease (CKD). Methods: We used homozygous TGR(mREN)27 rats (REN2), which develop severe high blood pressure around 4 weeks of age and heart failure at weeks of age. Six-week-old male REN2 rats were treated with a galectin-3 blocking compound, N-acetyllactosamine (Gal3i), for 6 weeks. Untreated REN2 and SD rats served as controls. We measured cardiac function with echocardiogram and invasive hemodynamics prior to sacrifice. Blood pressure and proteinuria were measured at 0, 3 and 6 weeks. Plasma creatinine was determined at 6 weeks. Renal damage was assessed using histological scores: focal glomerular sclerosis (FGS), glomerular desmin expression, glomerular and interstitial macrophage numbers and alpha smooth muscle actin expression. Inflammatory cytokines and extracellular matrix proteinases were quantified by RT-qPCR. Results: Systolic blood pressure was consistently higher in untreated REN2 rats compared with SD rats and was not affected by Gal3i treatment. Plasma creatinine and proteinuria were significantly increased in untreated REN2 rats and this was reduced by treatment with Gal3i. Parameters of renal damage were also elevated in untreated REN2 rats except for glomerular macrophage scores. All these parameters were reduced upon Gal3i treatment. Various inflammatory cytokines were elevated in untreated REN2 rats and attenuated by Gal3i. However, markers of extracellular matrix turnover were marginally altered in untreated REN2 rats as compared to SD rats. Conclusion: Pharmacological inhibition of galectin-3 attenuates hypertensive nephropathy, as indicated by a reduction in proteinuria, preservation of renal function and a decrease in renal damage. Drugs binding to galectin-3 may be potential therapeutic candidates for the prevention of chronic kidney disease (CKD). Keywords Chronic kidney disease, fibrosis, Galectin-3, renin-angiotensin system 96

98 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Introduction Renal impairment is frequently observed in cardiovascular disease (1, 2). Chronic kidney disease (CKD), defined as progressive loss of renal function over a period of month or years, is characterized at the level by glomerular sclerosis and interstitial fibrosis (5). CKD represents a significantly global health problem (3, 4). Early detection and prevention of CKD could improve both renal and cardiovascular morbidity and mortality. Glomerular sclerosis is an important factor in the progression of chronic kidney injury. Accordingly, drugs that aim to protect against glomerular injury can be of great value. Galectin-3 belongs to the galectin-3 family of mammalian lectins and is characterized by a carbohydrate recognition domain (CRD) that has affinity for β-galactosides. Galectin-3 mediates cell-cell and cell-matrix interactions by binding to lactosamine-containing glycoconjugates via its CRD (6). There is accumulating evidence that galectin-3 plays an important role in inflammatory and fibrotic processes (7). Upregulation of galectin-3 is involved in various types of organ fibrosis. Macrophage derived galectin-3 induces myofibroblast activation, promotes collagen synthesis, deposition and subsequently leads to fibrosis. Moreover, previous research showed that binding of modified citrus pectin (MCP) to the galectin-3 CRD exerts beneficial effects in experimental acute kidney injury (8). Transgenic TGR (mren2)27 (REN2) rats exhibit persistent high blood pressure, progressive proteinuria and nephropathy that strongly resembles the human situation: injury and dysfunction of glomerular endothelial cells, micro-inflammation, excessive production of extracellular matrix (ECM), which eventually results in glomerular sclerosis (9, 10). This collective of inflammation, glomerular sclerosis, tubular interstitial fibrosis, and proteinuria are all early markers for progressive renal dysfunction in CKD (11, 12) Previous experimental and clinical studies demonstrate that proteinuria caused by hypertension could almost be totally reversed by angiotensin converting enzyme (ACE) inhibitor (13, 14), whereas other experimental studies showed that reduction in proteinuria was partially independent of blood pressure level (15-17). In the present study, we examined the effects of pharmacological inhibition of galectin-3 by N-acetyllactosamine (N-lac, Gal3i) on progressive glomerulosclerosis and proteinuria in hypertensive REN2 rats. 5 Materials and methods Animals We studied 6-week-old, male, homozygous, TGR (mren2)27 rats (denoted as REN2). These rats show a phenotype of severe hypertension and left ventricular (LV) hypertrophy, leading to heart failure (HF) at weeks of age (18). Rats were bred at the MaxDelbrück Center for Molecular Medicine (Berlin, Germany). Male age-matched Sprague-Dawley (SD) rats were used as control (Harlan, The Netherlands). Animals were housed under standard conditions. The study was approved by the Animal Ethical Committee of the University of 97

99 Chapter 5 Groningen, the Netherlands, and was conducted in accordance with existing guidelines for the care and use of laboratory animals. Experimental design To study the effects of galectin-3 inhibition on renal protection, we allocated SD rats and REN2 rats to control treatment or treatment with the galectin-3 inhibitor. Rats were treated with an established inhibitor of galectin-3, N-Acetyllactosamine (N-Lac), in a final dose of 5 mg/kg/day. Injections were administered intraperitoneally, three times per week. Three different groups were studied: SD-control (N=5), Ren-2-control (N=5), Ren-2-Gal3-inhibitor (N=10). Rats were placed in metabolic cages during 24 hours at baseline and at 3 and 6 weeks and urine samples were collected for determination of urinary protein. Systolic blood pressure (SBP) was measured using a noninvasive tail-cuff method using a computer-assisted oscillatory detection device (Apollo 179; IITC Life Science, Woodland Hills, California, USA) at week 0, 3 and 6 after treatment, Rats were sacrificed 6 weeks after initiation of the experiment. Measurement of cardiac and hemodynamic function Cardiac function was assessed by echocardiography at baseline and prior to sacrifice (Vivid 7, GE Healthcare, Chalfont StGiles, UK; equipped with a 10-MHz (rats) phase array linear transducer, as described (19). Hemodynamic function was assessed invasively, as previously described (20), by introducing a 1.4 French microtip pressure-volume transducer (Millar Instr. Inc., Houston, TX, USA) via the right carotid artery into the aorta. A threeminute period was allowed for stabilization before systolic and diastolic blood pressure and heart rate (HR) were recorded (average of 20 heart cycles). Other parameters that were measured include peak systolic blood pressure (SBP) and LV end diastolic pressure (LVEDP). Tissue and plasma processing After measuring hemodynamics, arterial blood was drawn and collected. Samples were centrifuged at 3000 rpm for 15 min at 4 C, and plasma was frozen for creatinine analysis. Kidneys were removed and their weight was determined. Transversally cut kidneys were snapping frozen or fixed in buffered formalin (3.7%) for 24 hours, dehydrated and embedded in paraffin. Immunohistochemistry Paraffin sections were dewaxed and subjected to an antigen retrieval procedure by overnight incubation at 80 C in 0.1M Tris/HCL, ph 9.0. Sections were subsequently washed three times with PBS, endogenous peroxidase was blocked with 0.075% H 2 O 2 in phosphate buffered saline (PBS, ph 7.4) for 30 minutes and incubated with the following primary antibodies: the myofibroblast marker alpha smooth muscle actins, α-sma (clone 1A4, Sigma Aldrich, St.Louis, MO, USA); the macrophage marker ED1(#MCA341R, AbDSerotec, 98

100 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Oxford, UK); the tubular damage marker Kim-1 (a gift from Victoria Bailly, Principal Scientist, Cambridge, UK); and the glomerular epithelial marker desmin(dakopatts, DAKO, Golostrup, Denmark). All incubations with primary antibodies were in 1% BSA/PBS for 1 hour at room temperature. Binding was detected using sequential incubation with appropriate peroxidase labeled secondary and tertiary antibodies diluted in 1% BSA/PBS buffer and 1% normal rat serum for 30 minutes, peroxidase activity was developed by using 3,3 - diaminobenzidine tetrachloride for 10 min containing 0.03% H 2 O 2. Ultimately, counterstaining was performed using Mayer s hematoxilin. All sections were digitalized using a scanning system at 20x magnification (Nanozoomer 2.0-HT, Hamamatsu, Japan). α-sma and Desmin staining were quantified with Scanscope software (Aperio Technologies version 9, Vista, CA, USA). Kim-1 staining was similarly analyzed in the entire renal cortex. The number of both glomerular (50 glomeruli) and interstitial macrophages (30 interstitial fields) was manually counted by a blinded observer. Additionally, kidney sections were stained with periodic acid-schiff (PAS) and scored for focal glomerular sclerosis (FGS). FGS was scored positive if all of the following features were present: collapse of capillaries, mesangial matrix expansion, and adhesion of the glomerular visceral epithelium to Bowman s capsule. A score for the degree of affected glomeruli was applied as follows: unaffected glomeruli were scored as 0; 0-25% affected glomeruli was scored as 1, 25-50% affected was scored as 2, 50-75% affected was scored as 3 and if all glomeruli were positive (75-100%) for FGS, a score of 4 was given. The ultimate score (%) is obtained by dividing the total score by the number of glomeruli times one hundred (21). Quantitative real-time PCR cdna synthesis was performed using 0.5 µg total RNA (Quantitect Rev. Transcriptase kit, Qiagen, Venlo, The Netherlands) as described previously (19, 22). Quantitative real-time PCR (RT-qPCR) was performed using SYBR Green mix (Absolute SYBR Green ROX mix, Thermo Scientific, Breda, the Netherlands) on CFX384 Real-Time PCR Detection System (Bio-Rad Laboratories, Veenendaal, the Netherlands). All targets gene expression level of inflammation cytokines and extracellular matrix turnover proteins (a list of primers used for RT-qPCR showed in supplemental table 1) were evaluated under the same reaction conditions: 95 C for 15 minutes, then 36 cycles of 95 C for 15 seconds and 60 C for 30 seconds. Samples were analyzed with quantification software (Bio-Rad CFX Manager 1.6). mrna levels were expressed in relative units based on a standard curve obtained with serial dilutions of a calibrator cdna mixture. Gapdh expression was used to normalize all expression data. 5 Statistical analysis Data are expressed as mean ± SEM. Means were compared using one -way ANOVA with a Dunnett post hoc test, using untreated REN2 rats as a comparator. All calculations were made using SPSS computer software, (version 18 SPSS, ll, Chicago, USA). 99

101 Table 1. Baseline characteristics of rats SD-con (N=5) REN2-con (N=5) REN2-Gal3i (N=10) BW (g) 377±6* 329±14 347±26 Kidneyweight (mg/gr) 6.61±0.2* 8.98± ±0.3 SBP (week 0, mmhg) 146±8* 229±13 220±5 SBP (week 3, mmhg) 151±9* 227±7 232±4 SBP (week 6, mmhg) 140±11* 216±8 235±19 Proteinuria (week0, mg/24h) 14.27±0.9* 37.17± ±3.49# Proteinuria (week3, mg/24h) 27.83±2.77* 59.26± ±2.96* Proteinuria (week6, mg/24h) 26.35±1.11* 80.58± ±1.8* FS (%) 44±2* 34±1 41±3* LVEDP (mmhg) 4±1* 9±1 5±2* Gal3i: galectin-3 inhibitor; BW: body weight; Kidney weight: corrected for BW, mg/gr; SBP, systolic blood pressure before treatment; FS: Fractionalshortening; LVEDP: Left ventricular end diastolic pressure. * P<0.05 vs. REN2-con, # P= 0.06 REN2-Gal3i vs. REN2-con.

102 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Characteristics of REN2 rats REN2 rats developed severe hypertension (>200mmHg) at 6 weeks of age; SBP was significantly increased compared to SD rats at all-time points (0, 3, 6 weeks, Table 1 and Figure 1b). There was no difference in SBP between the untreated REN2 and Gal3i treated REN2 rats (217±7 vs. 235±6 mmhg, P=NS) (Figure1b). Ren2 rats had lower body weights compared to SD rats. Cardiac and hemodynamic measurements prior to sacrifice showed a significant decrease in fractional shortening (FS) and an increase in LVEDP in REN2 rats compared to SD rats (34±1% vs. 44±2% and 9±1 vs. 4±1 mmhg, P<0.05) (Table1). The untreated REN2 rats develop LV dysfunction. Treatment with Gal3i attenuated LV dysfunction by interfering with myocardial fibrogenesis (unpublished data) (Table1). Figure 1 Renal function in rats. Outline of the experimental protocol of the prevention studies in rats (a). Changes in systolic blood pressure of rats, as assessed by tail-cuff measurements at week 0, 3 and 6 (b). Proteinuria measurements at week 0, 3 and 6 (c). Plasma creatinine at sacrifice (week 6) (d). Calculated creatinine clearance at week 6 (e). * P<0.05 vs. REN2-con. 5 Renal function Renal function was impaired in REN2 rats, as shown by a marked increase in proteinuria, plasma creatinine and an associated decrease in renal clearance after 6 weeks of the treatment (proteinuria: 81±18 vs. 26±1 mg/24h; plasma creatinine: 45±5 vs. 26±2 µmol/l; renal clearance: 2.1±0.1 vs. 4.0±0.3 ml/24h, P<0.05) (Table 1, Figure 1c-e). Galectin-3 inhibition prevents progression hypertensive kidney injury 101

103 Chapter 5 Proteinuria was increased in the untreated REN2 rats compared with SD rats, but the increase was attenuated in the Gal3i treated REN2 rats (16±2 vs. 81±19 mg/24h in untreated REN2 rats, P<0.05) (Figure1c). Also, a marked increased level of plasma creatinine was found in untreated REN2 rats, and this was also attenuated after treatment with the Gal3i (27±2 vs. 45±5 µmol/l in the untreated REN2 rat, P<0.05) (Figure1d). Creatinine clearance was clearly decreased in the untreated REN2 rats compared with SD rats, which was attenuated after treatment with Gal3i (2.8±0.4 vs. 2.2±0.1 ml/24h in the untreated REN2 rat, P<0.05) (Figure1e). Galectin-3 inhibition protects renal damage In untreated REN2 rats, FGS was significantly higher as compared to SD rats (15 ±4% vs. 0 % in SD rats, P<0.05). Gal3i treatment resulted in a significant attenuation of FGS (3 ±1% vs. 15±4% in untreated REN2 rat, P<0.05) (Figure 2a, b, c and g). Additionally, glomerular desmin expression was markedly increased in untreated REN2 rats (21±1 vs. 5±1 in SD rats, P<0.05), which was attenuated after Gal3i treatment (14±2 vs. untreated REN2 rat, P<0.05) (Figure 2d, e, f and h). Galectin-3 inhibition suppresses renal inflammatory response Persistent systemic hypertension induces glomerular micro-inflammation (9). We observed that the number of interstitial macrophages was significantly increased in the untreated REN2 rats as compared to SD rats, while the number of glomerular macrophages Figure 2 Glomerular morphology in the rat kidney. Representative pictures of focal glomerular sclerosis with PAS staining are shown for SD, untreated REN2, and Gal3i-treated REN2 rats (a-c). Glomerular desmin expression in SD, untreated REN2, and Gal3i-treated REN2 rats (d-f). The quantified percentages of focal glomerular sclerosis (g) and desmin expression (h). * P<0.05 vs. REN2-con 102

104 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Figure 3 Inflammatory cells and markers in rat kidney. Representative pictures of glomerular macrophages in SD, untreated REN2, and Gal3i-treated REN2 rats are shown (a-c). Typical examples of macrophage infiltration in the tubulointerstitial space of SD, untreated REN2, and the Gal3i-treated REN2 rats (d-f). Quantification of the number of glomerular macrophages (g) and macrophages in tubulointerstitial space (h). Inflammatory markers were quantified by RT-qPCR and corrected for Gapdh expression. Fold differences are shown for: Galectin-3 (i), Cd68 (j), Interleukin-6 (K) and Mcp- 1 (l). * P<0.05 vs. REN2- con 5 was not changed in the untreated REN2 group (Figures 3a, 3b, 3d, 3e, 3g, and 3h). Gal3i treatment attenuated macrophage influx in both the glomerular and the interstitial compartment (Figures3c, 3f, 3g and 3h). CD68 (macrophages), interleukin-6 (Il-6) and other relevant inflammatory cytokines e.g. galectin-3 and monocyte chemoattractant protein 1 (Mcp-1) were subsequently determined by RT-qPCR (Figure3i-l). Galectin-3, Il-6 and Mcp-1 were all significantly increased in untreated REN2 rats and reduced by the treatment with 103

105 Chapter 5 Figure 4 Renal tubular and interstitial changes in rat kidney. Representative pictures of tubulo-interstitial α-sma staining in SD, untreated REN2, and Gal3i-treated REN2 rats (a-c). Typical examples of tubular damage marker KIM-1 staining in SD, untreated REN2, and Gal3i-treated REN2 rats (d-f). Quantification of α- SMA staining (g) and KIM-1 mrna expression (h). * P<0.05 vs. REN2-con Gal3i. Cd68 mrna level was not altered in the untreated REN2 group as compared to the SD group. Gal3i treatment did, however, attenuate Cd68 expression, which is in line with the decreased glomerular and interstitial influx of macrophages in this animal group. Matricellular protein changes and in early stages of kidney injury α-sma expression, as investigated by IHC and RT-qPCR was significantly increased in untreated REN2 rats as compared to SD rats. Gal3i caused a reduction in α-sma IHCstaining and mrna levels (Figures 4a-c, 4g and supplemental figure 1a). Also, a decrease in KIM-1 protein expression was observed after Gal3i treatment (Figures 4d-f) and Kim-1 mrna expression was similarly decreased (Figure 4h). Additionally, the mrna levels of Tgf-β, Mmp2 and Timp2 were not altered in the untreated REN2 group as compared to the SD group (Supplemental figures 1b, 1c and 1f). Interestingly, Mmp9 tends to show a lower expression in both untreated and treated REN2 groups as compared to the SD group (Supplemental figure 1d). Furthermore, Timp1 expression was significantly increased in the untreated REN2 group and decreased when treated with Gal3i (Supplemental figure 1e). Finally, we found that mrna expression of extracellular matrix collagen proteins Col1a1 and Col3a1 were not different between the untreated REN2 and the SD groups, but significantly decreased when treated with Gal3i (Supplemental figures 1g and 1h). 104

106 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Discussion The major finding of the present study study is that targeted inhibition of galectin-3 attenuates renal structural and functional deterioration in REN2 rats with hypertensive end organ damage and increased plasma galectin-3 levels. Over a six weeks course, the untreated REN2 rats developed substantial proteinuria which was associated with the development of glomerulosclerosis. Treatment with Gal3i almost completely prevented the development of proteinuria and attenuated fibrogenesis and inflammation, as determined by histological staining and gene expression analysis. Interestingly, hypertension is not altered in REN2 rats treated with the Gal3i, indicating that the positive effects of the treatment are blood pressure independent. Galectin-3 is expressed in various organs, including heart, lung, liver and kidney (7). The role of galectin-3 in fibrosis and inflammation has been partially elucidated in recent years. In a mouse model of unilateral ureter obstruction (UUO), characterized by severe hydronephrosis, inflammation and fibrosis galectin-3 disruption (Gal-3 KO mice) resulted in decreased renal inflammation. This was associated with a reduced pro-fibrotic response as evidenced by a decreased collagen production and deposition (23). Another study in UUO-induced renal injury also reported reduced myofibroblast activation as determined by decreased α-sma staining in Gal-3 KO mice as compared to wild type mice. Surprisingly, they observed increased interstitial collagen deposition in these mice and decreased E-cadherin levels, a surrogate marker for tubular damage (24). Unfortunately, no explanation is given for this paradoxical result of less interstitial myofibroblasts and less procollagen I expression, but more fibrosis (collagen deposition). These remarkable effects may occur at the late stages (2-3 weeks induction) of the UUO model, since Henderson et al., who investigated an earlier time point (7 days) after UUO-induced kidney injury observed both less myofibroblasts and less fibrosis in Gal-3 KO mice. In a model of folic acid induced kidney fibrosis it was shown that treatment with modified citrus pectin (MCP), a compound that binds to the galectin-3 carbohydrate recognition domain (CRD), attenuated macrophage influx and renal fibrosis (8). Together with our results, which reveal Gal3i mediated attenuation of inflammatory responses (e.g. CD68, IL-6 and MCP-1), macrophage infiltration and myofibroblast activationin a hypertensive rat model, there is strong support for a protective role for galectin-3 inhibition in different models of renal damage. Future studies require the analysis of long-term treatment effects and the effects of Galectin-3 inhibition in established kidney disease. TGF-β is an important mediator of myofibroblast activation (25, 26) and it was demonstrated that macrophage derived galectin-3 directly induces myofibroblast activation resulting in up-regulation of collagen synthesis (23, 27). We did not observe altered renal expression of Tgf-β mrna in our hypertensive rat model and Gal3i treatment did also not affect TGF-β expression. This is in line with previous observations (23). This suggests that in our model TGF-β may not be the main driving force for fibrogenesis and that galectin-3 does not affect the expression of TGF-β. Gal3i did, however, lower the expression of several other genes involved in extracellular matrix (ECM) remodeling, including Timp1 and Col1a1 and 5 105

107 Chapter 5 Col3a1. The expression of the metalloproteases Mmp2 and Mmp9 was not altered by Gal3i treatment, indicating that galectin-3 inhibition alters the expression of a limited number of ECM genes independent of TGF-β modulation. Numerous studies have shown that anti-hypertensive therapy can slow down the decline in renal dysfunction, (28, 29), although other experimental studies have shown that a reduction in proteinuria can be mediated independent of blood pressure. Intervention with tranilast, an inhibitor of TGF-β, shows beneficial effects on proteinuria and tubulointerstitial damage independent of blood pressure in streptozotocin induced diabetic REN2 rats (15). Also, statin treatment reduces glomerular inflammation and podocyte damage in experimental deocycoricosterone-acetate (DOCA)-salt hypertensive rats (16). Moreover, arrest-specific protein 6 (Gas6) is involved in cardiac and renal injury, and Gas6 deficiency reduced renal inflammation, fibrosis and cardiac remodeling independent of blood pressure (17). Collectively, the above mentioned studies, together with our galectin-3 inhibitory study indicate that renal protection can be reached not only by lowering blood pressure levels, but also by specific anti-fibrotic and anti-inflammatory treatments. Conclusion Our study shows that pharmacological inhibition of galectin-3 attenuates impaired CKD in a hypertensive rat model. Galectin-3 inhibition attenuates myofibroblast activation and inflammation resulting in reduced fibrogenesis. This improves glomerular filtration function and reduces proteinuria. Therefore, we conclude that galectin-3 inhibition exerts its protective effects by directly acting on the renal glomeruli, parenchyma and tubuli and is independent of blood pressure. These new findings warrant further studies using galectin-3 inhibition as a potential novel supplementation to CKD therapy. 106

108 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy References 1. Smith GL, Lichtman JH, Bracken MB, et al. Renal impairment and outcomes in heart failure: Systematic review and metaanalysis. J Am Coll Cardiol 2006;47(10): Hillege HL, Girbes AR, de Kam PJ, et al. Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation 2000;102(2): Klahr S, Levey AS, Beck GJ, et al. The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease. modification of diet in renal disease study group. N Engl J Med 1994;330(13): Jafar TH, Stark PC, Schmid CH, et al. Progression of chronic kidney disease: The role of blood pressure control, proteinuria, and angiotensin-converting enzyme inhibition: A patient-level meta-analysis. Ann Intern Med 2003;139(4): Becker GJ, Hewitson TD. The role of tubulointerstitial injury in chronic renal failure. Curr Opin Nephrol Hypertens 2000;9(2): Ochieng J, Furtak V, Lukyanov P. Extracellular functions of galectin-3. Glycoconj J 2004;19(7-9): Yang RY, Rabinovich GA, Liu FT. Galectins: Structure, function and therapeutic potential. Expert Rev Mol Med 2008;10:e Kolatsi-Joannou M, Price KL, Winyard PJ, et al. Modified citrus pectin reduces galectin-3 expression and disease severity in experimental acute kidney injury. PLoS One 2011;6(4):e Meguid El Nahas A, Bello AK. Chronic kidney disease: The global challenge. Lancet 2005;365(9456): El-Nahas AM. Plasticity of kidney cells: Role in kidney remodeling and scarring. Kidney Int 2003;64(5): Granger JP. An emerging role for inflammatory cytokines in hypertension. Am J Physiol Heart Circ Physiol 2006;290(3):H Hall JE, Crook ED, Jones DW, et al. Mechanisms of obesity-associated cardiovascular and renal disease. Am J Med Sci 2002;324(3): Ibsen H, Wachtell K, Olsen MH, et al. Albuminuria and cardiovascular risk in hypertensive patients with left ventricular hypertrophy: The LIFE study. Kidney Int Suppl 2004;(92)(92):S Brouwers FP, de Boer RA, van der Harst P, et al. Influence of age on the prognostic value of mid-regional proadrenomedullin in the general population. Heart 2012;98(18): Mifsud S, Kelly DJ, Qi W, et al. Intervention with tranilast attenuates renal pathology and albuminuria in advanced experimental diabetic nephropathy. Nephron Physiol 2003;95(4):p Hartner A, Klanke B, Cordasic N, et al. Statin treatment reduces glomerular inflammation and podocyte damage in rat deoxycorticosterone-acetate-salt hypertension. J Hypertens 2009;27(2): Park JK, Theuer S, Kirsch T, et al. Growth arrest specific protein 6 participates in DOCA-induced target-organ damage. Hypertension 2009;54(2): Ruifrok WP, Qian C, Sillje HH, et al. Heart failure-associated anemia: Bone marrow dysfunction and response to erythropoietin. J Mol Med (Berl) 2011;89(4): Kuipers I, van der Harst P, Kuipers F, et al. Activation of liver X receptor-alpha reduces activation of the renal and cardiac renin-angiotensin-aldosterone system. Lab Invest 2010;90(4):

109 Chapter Pacher P, Nagayama T, Mukhopadhyay P, et al. Measurement of cardiac function using pressure-volume conductance catheter technique in mice and rats. Nat Protoc 2008;3(9): Hamming I, Navis G, Kocks MJ, et al. ACE inhibition has adverse renal effects during dietary sodium restriction in proteinuric and healthy rats. J Pathol 2006;209(1): Lu B, Tigchelaar W, Ruifrok WP, et al. DHRS7c, a novel cardiomyocyte-expressed gene that is down-regulated by adrenergic stimulation and in heart failure. Eur J Heart Fail 2012;14(1): Henderson NC, Mackinnon AC, Farnworth SL, et al. Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol 2008;172(2): Okamura DM, Pasichnyk K, Lopez-Guisa JM, et al. Galectin-3 preserves renal tubules and modulates extracellular matrix remodeling in progressive fibrosis. Am J Physiol Renal Physiol 2011;300(1):F Sharma K, Ziyadeh FN. The emerging role of transforming growth factor-beta in kidney diseases. Am J Physiol 1994;266(6 Pt 2):F Sato M, Muragaki Y, Saika S, et al. Targeted disruption of TGF-beta1/Smad3 signaling protects against renal tubulointerstitial fibrosis induced by unilateral ureteral obstruction. J Clin Invest 2003;112(10): Sharma UC, Pokharel S, van Brakel TJ, et al. Galectin-3 marks activated macrophages in failure-prone hypertrophied hearts and contributes to cardiac dysfunction. Circulation 2004;110(19): Witte K, Schnecko A, Schmidt T, et al. Cardiovascular risk, renal hypertensive damage, and effects of amlodipine treatment in transgenic TGR(mREN2)27 rats. Gen Pharmacol 1999;33(5): Peterson JC, Adler S, Burkart JM, et al. Blood pressure control, proteinuria, and the progression of renal disease. the modification of diet in renal disease study. Ann Intern Med 1995;123(10):

110 Pharmacological Inhibition of Galectin-3 Protects Against Hypertensive Nephropathy Supplemental Table S1. List of primers used for RT-qPCR in rats renal RT-qPCR primer, 5 to 3 Gene symbol (name) Forward Reverse Lgals3 (Galectin-3) CCCGCTTCAATGAGAACAAC ACCGCAACCTTGAAGTGGTC α-sma CATCATGCGTCTGGACTTGG TCACGCTCAGCAGTAGTCAC Cd-68 CTCTCATCATTGGCCTGGTC GGGCTGGTAGGTTGATTGTC Il-6 (Interleukin-6) CCCACCAGGAACGAAAGTCA TCTTGCGGAGAGAAACTT Mcp-1 CCGACTCATTGGGATCATCTT TGTCTCAGCCAGATGCAGTTAAT Kim-1 AGAGAGAGCAGGACACAGGCTT ACCCGTGGTAGTCCCAAACA Tgf-β AAGAAGTCACCCGCGTGCTA TGTGTGATGTCTTTGGTTTTGTCA Col1a1 ACAGCGTAGCCTACATGG AAGTTCCGGTGTGACTCG Col3a1 TGGAAACCGGAGAAACATGC CAGGATTGCCATAGCTGAAC Mmp2 TGAGCTCCCGGAAAAGATTG CATTCCCTGCGAAGAACACA Mmp9 CGGGAACGTATCTGGAAATTCG CATGGCAGAAATAGGCCTTGTC Timp1 AGAGCCTCTGTGGATATGTC CTCAGATTATGCCAGGGAAC Timp2 TGGACGTTGGAGGAAAGAAG TGTCCCAGGGCACAATAAAG Gapdh CATCAAGAAGGTGGTGAAGC ACCACCCTGTTGCTGTAG Figure S1 Gene expression of ECM proteins changed in rat kidney. α-sma(a), Tgf-β(b), Mmp2(c), Mmp9(d), Timp1(e), Timp2(f), Col1a1(g), Col3a1(h). * P<0.05 vs. REN2-con 5 109

111

112

113

114 Chapter 6 Clinical Correlations of Plasma Galectin-3 Levels in a Well-defined Chronic Heart Failure Cohort Lili Yu; Willem P.T. Ruifrok; Ysbrand T. Klip Peter van der Meer MD, PhD; Wiek H. van Gilst Rudolf A. de Boer Manuscript

115 Chapter 6 Abstract Aims - Galectin-3 plays an important role in fibrogenesis. Furthermore, increased galectin-3 levels are associated with poor survival in patients with heart failure (HF). We examined the correlation of plasma galectin-3 levels with cardiopulmonary aerobic capacity and renal function in patients with chronic HF. Methods and results - We measured plasma galectin-3 in 99 patients with stable chronic HF with New York Heart Association (NYHA) class II-IV. All patients had left ventricular ejection fraction (LVEF) 45% and an ability to undergo cardiopulmonary exercise testing. In the present HF cohort, plasma galectin-3 levels were divided in quartiles (quartile 1: <12.65 ng/ml; quartile 2: ng/ml; quartile 3: ng/ml; quartile 4: >18.67 ng/ml). High galectin-3 levels were associated with poor renal function (consisted of increased creatinin (p=0.026); increased urea (p=0.01); decreased egfr (p=0.01), increased NT-proBNP (p=0.008), and decreased peak VO 2 (p=0.038). Linear regression analysis showed a correlation between the plasma galectin-3 levels and peak oxygen uptake (VO 2 max), p=0.016; and renal function (p=0.002). However, after adjustment for age and gender, the correlation between galectin-3 and VO 2 max and renal function was lost. Conclusions - high plasma galectin-3 levels are associated with poor renal function and lower aerobic capacity in patients with chronic HF. Key words Chronic heart failure, Galectin-3, renal function, VO 2 max, cardiopulmonary aerobic capacity. 114

116 Clinical Correlations of Plasma Galectin-3 Levels in a Well-defined Chronic Heart Failure Cohort Introduction Heart failure (HF) is a serious medical disease and an epidemiological problem. It is characterized by high morbidity and mortality [1-3]. The pathophysiologic mechanisms of HF appear to be the results of interaction between cardiac remodeling, neurohormonal peptides (e.g. N-terminal pro brain natriuretic peptide (NT-proBNP)), inflammation, and different biomarkers [4-6]. Accumulated experimental studies reported that macrophage-derived galectin-3 plays important regulatory roles in inflammation and fibrotic processes in the development cardiac remodeling and chronic HF [4-6]. Additionally, clinical evidence showed that plasma galectin-3 levels are increased in patients with acute and chronic HF [7-14]. The PRIDE study revealed that plasma galectin-3 was a superior predictor for 60-day mortality compared to NT-proBNP [11]. Subsequently, high plasma galectin-3 levels were associated with left ventricular filling and diastolic function [7]. Furthermore, in patients with chronic stable and acute decompensated HF, increased plasma galectin-3 levels were linked to renal dysfunction and lower peak oxygen uptake (VO 2 max) [8, 11, 15]. In the DEAL-HF study, plasma galectin-3 levels were increased in patients with higher NT-proBNP levels, which were in turn correlated with lower estimated glomerular filtration rate (egfr) and lower VO 2 max [13]. Furthermore, higher levels of galectin-3 were found in patients with renal dysfunction as compared to patients with normal renal function. As a substantial part of chronic HF patients have decreased renal function, we examined the correlation of plasma galectin-3 levels on cardiopulmonary aerobic capacity and renal function in patients with chronic HF. Methods Patients and study design The data described and used in this manuscript is derived from the BENEFICIAL study (Effects of alagebrium, an advanced glycation end product breaker, on exercise tolerance and cardiac function in patients with chronic heart failure, NTC ) [16-18]. The study design, baseline characteristics, inclusion and exclusion criteria have been published previously [16-18]. Data of all 99 patients who were recruited from the University Medical Center Groningen, the Netherlands and three other regional affiliated hospitals were analyzed in this sub study. Briefly, patients with New York Heart Association (NYHA) class II-IV had to have stable chronic HF for at least three months, and a documented left ventricular ejection fraction (LVEF) <45%. Main exclusion criteria were the inability of patients to undergo exercise testing, cardiac resynchronization therapy, pacemaker therapy, active and/or treated malignancies within 12 months prior to inclusion, and clinically significant renal dysfunction. The efficacy measurements included echocardiography and cardiopulmonary aerobic capacity testing. The BENEFICIAL study was approved by the Medical Ethical Committee of the University Medical Center Groningen and all subjects gave written informed consent

117 Chapter 6 Echocardiography Two-dimensional echocardiography was performed by experienced cardiac technicians using a General Electric VIVID 7 system with a MHz probe (Horton, Norway). Left ventricular dimensions were measured. Diastolic function was evaluated with peak early (E) and late (A) diastolic filling velocities, isovolumetric relaxation time (IVRT) and deceleration time (Dct) of the early peak filling. Early diastolic tissue velocity (E') was measured on the lateral and septal wall areas, using color-coded tissue Doppler imaging (CC-TDI). E/E' was calculated by dividing the peak early diastolic filling (E) by the average E'. Diastolic dysfunction was defined as an E/E' >10. Systolic dysfunction was determined by Simpson s LVEF and defined as a LVEF 45%. If Simpson s LVEF could not be determined, LVEF was estimated by eyeballing [16-18]. Cardiopulmonary aerobic capacity testing Cardiopulmonary aerobic capacity testing was performed using a care fusion, Master screen CPX (Houten, The Netherlands) according to a modified Bruce protocol [19], which increases the workload more gradually than the Bruce protocol [20]. The first stage was performed at 1.7 mph and 0% grade, the second stage at 1.7 mph and 5% grade, and the third stage corresponds to the first stage of the Bruce protocol. A standard 12 lead electrocardiogram was recorded continuously during exercise testing. Blood pressure was registered on a regulatory basis using a manual cuff sphygmomanometer. Patients were encouraged to continue the exercise until their peak oxygen uptake (denoted as VO 2 max) was reached or when they became symptomatic, or discontinuation was indicated for safety reasons. Oxygen uptake, carbon dioxide production, and minute ventilation were measured using breath-by-breath gas analysis. Peak VO 2 was determined as an average value of the two highest VO 2 values at peak performance, data were expressed as ml/kg/min [16-18]. Biochemical measurements Plasma Galectin-3 levels were determined by an enzyme-linked immunosorbent assay (ELISA) developed by BG Medicine (Galectin-3 assay TM, BG Medicine, Inc., Waltham, USA). The assay quantitatively measures the concentration of human galectin-3 levels in EDTA plasma. This assay has a high sensitivity (lower limit of detection 1.13 ng/ml) and exhibits no cross-reactivity with collagens or other members of the galectin family. Calibration of the assay was performed according to the manufacturer s recommendation and values were normalized to a standard curve [9, 10]. NT-proBNP levels were measured by an immunoelectro-chemiluminesence method (Elecsys, Roche Diagnostics, Basel, Switzerland) [21]. The egfr was estimated using the simplified modification of Diet in Renal Disease (smdrd) formula [22]. Statistical analysis 116

118 Clinical Correlations of Plasma Galectin-3 Levels in a Well-defined Chronic Heart Failure Cohort Data are expressed as mean ± standard deviation (SD) when normally distributed. Data are expressed as medians with lower and upper quartiles when non-normally distributed. Categorial variables are expressed as frequencies and percentages. Baseline characteristics were divided into quartiles of plasma galectin-3 levels. Differences between groups were compared using the 1-way analysis of variance test, Kruskal-Wallis test or Chisquare test where appropriate. For further analyses, logarithmic transformation was performed to achieve a normal distribution for skewed variables. The Pearson correlation coefficient was employed to test correlations between galectin-3 and other variables. All tests were two-sided and a p-value <0.05 was considered statistically significant. All statistical analyses were performed using STATA version 11.0 (StataCorp LP) and SPSS version 18.0 (SPSS Inc). Results Patient characteristics Baseline characteristics of all patients, according to quartiles of plasma galectin-3, are described in table 1. Overall, mean age of the study population was 61±11 years, and 80% were males. Around 40% patients had NYHA-class III and IV. Mean LVEF was 32±9%. Mean egfr was 80±21 ml/min/1.73m 2, mean NT-proBNP value was 388 ( ) ng/l, mean VO 2 max was 21.7±6.1 ml/kg/min. All patients were on standard medication for HF, including an ACE inhibitor (ACEi) or angiotensin II receptor blocker (ARB), beta-blocker (BB), and diuretics. Galectin-3 and parameters of disease severity and renal function Plasma galectin-3 levels displayed a moderately significant correlation with levels of NTproBNP (R=0.26; p=0.009; figure 1A), but not with NYHA-class (R=0.13; p=0.230). Furthermore, galectin-3 levels also showed a correlation with creatinin levels (R=0.24; p=0.014) and with levels of plasma urea (R=0.33; p=0.001). In addition, linear regression analysis showed a significant association between plasma galectin-3 levels and egfr (R=- 0.30; p=0.002; figure 1B). When adjusted for age and gender plasma galectin-3 levels are not correlated with egfr, suggesting that some of the prognostic power of galectin-3 may be associated with age and gender. 6 Correlation between plasma galectin-3 levels and exercise capacity At baseline, VO 2 max correlated with increasing galectin-3 levels (R=-0.24; p=0.016; figure 1C). In contrast, the level of plasma galectin-3 is not associated with the resting oxygen uptake (R=-0.11; p=0.285). Furthermore, linear regression analysis showed a correlation between the plasma galectin-3 levels and the VO 2 max (p=0.016). However, when corrected for age and gender, plasma galectin-3 is no longer correlated with VO 2 max (figure 1). Correlation between plasma galectin-3 and echocardiographic parameters 117

119 Table 1. Baseline parameters according to the plasma galectin-3 levels Variables All patients (n=99) Quartile 1 (<12.65 ng/ml) Quartile 2 ( ng/ml) Quartiles of galectin-3 Quartile 3 ( ng/ml) Quartile 4 (>18.67 ng/ml) p-value for trend Age, years ACEi ± ± ± 9 63 ± ± Sex, male, % Ischemic cause of HF, % BMI, kg/m 2 28±4 28±3 28±4 28±5 27± NYHA class, %, II/III/IV 64/33/3 71/29/0 58/40/2 72/28/0 56/36/ SBP, mmhg 115±15 113±12 115±14 116±17 115± DBP, mmhg 72±9 73±9 71±8 75±11 70± Heart rate, bpm 69±14 71±17 64±14 72±11 71± Co-morbidities (%) Hypertension Hypercholesterolemia Diabetes Laboratory Hb, mmol/l 8.9± ± ± ± ± Sodium, mmol/l 140±2 140±2 140±2 140±2 141± Potassium, mmol/l 4.2± ± ± ± ± Total cholesterol, mmol/l 4.4± ± ± ± ± Ferritin, g/l 128 (71-206) 117 (59-296) 173 ( ) 126 (86-181) 111 (49-150) TSAT, % 31 (25-39) 32 (23-40) 32 (27-39) 35 (28-41) 27 (19-36) stfr, mg/l 1.21 ( ) 1.14( ) 1.21 ( ) 1.22 ( ) 1.25 ( ) Creat, mol/l 93±25 90±20 83±18 93±28 104± Urea, mmol/l 7.6± ± ± ± ± egfr, ml/min/1.73m 2 80±21 84±13 87±21 80±22 69± NT-proBNP, ng/l 388( ) 312( ) 199(94-380) 463( ) 499( ) hs-crp, mg/l 1.6 ( ) 1.4 ( ) 1.3 ( ) 2.1 ( ) 2.1 ( ) Treatment (%)

120 ARB BB Diuretics Exercise test Resting VO2, ml/min 357±81 370±86 346±71 377±93 337± VO2max, ml/kg/min 21.7± ±5.8 23± ± ± RQ 1.06± ± ± ± ± Echo parameters LVEDD, mm 59(54-62) 60(55-62) 58(48-62) 59(57-62) 57(53-63) LVESD, mm 47(41-53) 59(42-53) 46(41-52) 47(43-55) 44(39-54) E/A ratio 0.89 ( ) 0.84( ) 0.93 ( ) 0.88 ( ) 0.90 ( ) E/E' ratio 12.6 ( ) 11.7 ( ) 12.8 ( ) 12.7 ( ) 14.6 ( ) IVRT, ms 99 (83-116) 94 (89-118) 105 (89-126) 96 (83-107) 95 (83-107) Dct, ms 209 ( ) 219 ( ) 195 ( ) 190 ( ) 210 ( ) TAPSE 19 (17-23) 20 (18-22) 19 (18-23) 19 (15-25) 20 (17-23) Simpson LVEF, % 32±9 34±9 34±8 32±6 29± Eyeballing LVEF, % 32±10 33±9 34±9 30±9 31± HF: heart failure; BMI: body mass index; NYHA: New York Heart Association; SBP: systolic blood pressure; DBP: diastolic blood pressure; bpm: beats per minute; Hb: hemoglobin; TSAT: transferrin saturation; stfr: serum transferrin receptor; creat: creatinin; egfr: estimated glomerular filtration rate; NT-proBNP: N-terminal pro brain natriuretic peptide; hs-crp: high-sensitive C-reactive protein; ACEi: ACE-inhibitor; ARB: angiotenis II receptor blocker; BB: beta-blocker; RQ: respiration quotient; LVEDD: left ventricular end diastolic diameter; LVESD: left ventricular end systolic diameter; IVRT: isovolumetric relaxation time; Dct: deceleration time; LVEF: left ventricular ejection fraction; E': early diastolic tissue velocity; E; peak early filling velocity; A: late diastolic filling velocity.

121 Chapter 6 We analyzed the relation between plasma galectin-3 and different echocardiographic parameters. The interactions are described in table 1. The results show that increased plasma galectin-3 levels are not associated with parameters of diastolic function (E/A, p=0.872; E/E', p=0.351; IVRT, p=0.332; Dct, p=0.662) or systolic function (Simpson LVEF%, p=0.373) (table 1). Linear regression analysis showed that there is no correlation between plasma galectin-3 and LVEF (R=-0.21; p=0.105; figure 1D). A B egfr (ml/min/1.73m3) R = -0.30; P = Log NT-proBNP (ng/ml) R = 0.26; P = Log galectin Log Galectin-3 C D VO2max (ml/kg/min) R = -0.24; P = LVEF (%) R = -0.21; P = Log galectin Log galectin-3 Figure 1 Univariate relation (and 95% confidence intervals) between log galectin-3 in HF patients and renal function (A), NT-proBNP (B), VO 2 max (C) and LVEF (D) Discussion The present data shows that if galectin-3 levels in patients with chronic HF are high at baseline, this is strongly associated with increased creatinin, increased urea and decreased egfr. Furthermore, high galectin-3 levels at baseline are associated with increased levels of NT-proBNP. However, when corrected for age and gender, plasma galectin-3 shows no significant relation with egfr. 120

122 Clinical Correlations of Plasma Galectin-3 Levels in a Well-defined Chronic Heart Failure Cohort We observed a strongly correlation between plasma galectin-3 and VO 2 max, however, this correlation is abrogated after correction for age and gender. Finally, no correlation was found between increased levels of galectin-3 and echocardiographic measurements of cardiac function in chronic HF. Galectin-3 was discovered around ten years ago. It is widely distributed throughout the entire body, including heart, lung, liver and kidney [23]. The role of galectin-3 in fibrosis and inflammation has been elucidated in recent years. The first experimental evidence showing involvement of galectin-3 in chronic HF stems from a landmark study of Sharma and colleagues [4]. They demonstrated that galectin-3 could be used as a new target for intervention in chronic HF. Since then, clinical trials have consistently shown potential clinical usefulness of galectin-3 as a prognostic biomarker for chronic HF. Herein, van Kimmenade et al. were the first to evaluate the prognostic and predictive value of galectin-3 as a biomarker in acute and chronic HF [11]. The PRIDE study revealed that plasma galectin-3 was a superior predictor when compared to NT-proBNP. High plasma galectin-3 levels are associated with left ventricular filling and decreased diastolic function [7]. In contrast, in our study we found no relation between plasma galectin-3 levels and echocardiographic parameters for diastolic and systolic function in this cohort of patients with chronic HF. We argue that the discrepancy between previous reports and our study is due to fact that the patient s population in the PRIDE study is different from our present study (acutely decompensated HF in PRIDE vs. well defined stable chronic HF in the present study). In the DEAL-HF study, plasma galectin-3 levels were increased in patients with higher NT-proBNP levels. Furthermore high galectin-3 levels were associated with decreased egfr. Additionally, in the PRIDE and the PREVEND studies, our group showed a correlation between increased plasma galectin-3 levels and cardiovascular risk factors and renal dysfunction. Notably, growing evidence shows that renal dysfunction is frequently observed in cardiovascular disease [25, 26], being one of the most powerful predictors in chronic HF prognosis and plays an important role in the pathophysiologic process of HF [27]. Taken together, evidence is accumulating that increased plasma galectin-3 levels are associated with cardiovascular disease and renal dysfunction. We confirm these observations in our present study. Interestingly, in a recently published paper by Gopal et al. plasma galectin-3 is inversely related to renal function in patients with and without clinical HF [28]. They showed that galectin-3 correlated strongly with egfr, both in patients with HF and in patients without HF, and this relationship was unaffected by the presence or absence of clinical HF. They concluded that concentrations of plasma galectin-3 do not seem to depend on the level of compensation or type of HF. Furthermore, the relationship between galectin-3 and renal function seems to be affected little or not at all by the presence or absence of clinical HF. Cardiopulmonary aerobic capacity testing is one of the diagnostic tools for chronic HF. Decreased peak exercise capacity is associated with poor prognostic and decreased patient 6 121

123 Chapter 6 survival. In the HF-ACTION study, plasma galectin-3 levels were measured in 895 subjects with chronic HF from a randomized controlled trial of exercise training in patients with chronic HF (NYHA class II, III or IV). Galectin-3 was associated with increased NYHA-class, and lower VO 2 max [13]. The present data confirms that increased plasma galectin-3 levels are associated with decreased VO 2 max. Some limitations apply to this study. This is a sub study of the BENEFICIAL study and the BENEFICIAL study was not powered for the current analyses. After correction for age and gender, plasma galectin-3 shows no significant relation with egfr, VO 2 max and echocardiographic measurements of cardiac function, where previous studies do. This is probably due to the low number of subjects participating in this study. As a multifunctional biomarker, galectin-3 promotes macrophage migration, myofibroblast activation and collagen synthesis, all involved in organ fibrogenesis process. The relationship of galectin-3 with other cardiovascular markers (e.g.: LVEF, NT-proBNP), renal function and VO 2 max suggests a role of galectin-3 in integrating these mechanisms in the progression of HF. However, still much is unknown about the role of galectin-3 in cardiovascular disease. Well-designed studies are needed to further elucidate its role in chronic HF. Conclusions This study demonstrates that high plasma galectin-3 levels are correlated with poor renal function and lower aerobic capacity in patients with chronic HF. No significant relation was observed between plasma galectin-3 levels and echocardiographic parameters for chronic HF. Although much about the details of galectin-3 in HF remains vague, continued efforts at increasing precision may uncover a new chapter in our understanding of HF pathophysiology. 122

124 Clinical Correlations of Plasma Galectin-3 Levels in a Well-defined Chronic Heart Failure Cohort References 1. McMurray JJ, Teerlink JR, Cotter G, Bourge RC, Cleland JG, Jondeau G, Krum H, Metra M, O'Connor CM, Parker JD, Torre-Amione G, van Veldhuisen DJ, Lewsey J, Frey A, Rainisio M, Kobrin I, VERITAS Investigators (2007) Effects of tezosentan on symptoms and clinical outcomes in patients with acute heart failure: the VERITAS randomized controlled trials. JAMA 298: Jaarsma T, van der Wal MH, Lesman-Leegte I, Luttik ML, Hogenhuis J, Veeger NJ, Sanderman R, Hoes AW, van Gilst WH, Lok DJ, Dunselman PH, Tijssen JG, Hillege HL, van Veldhuisen DJ, Coordinating Study Evaluating Outcomes of Advising and Counseling in Heart Failure (COACH) Investigators (2008) Effect of moderate or intensive disease management program on outcome in patients with heart failure: Coordinating Study Evaluating Outcomes of Advising and Counseling in Heart Failure (COACH). Arch Intern Med 168: Dickstein K, Cohen-Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, Stromberg A, van Veldhuisen DJ, Atar D, Hoes AW, Keren A, Mebazaa A, Nieminen M, Priori SG, Swedberg K, ESC Committee for Practice Guidelines (CPG) (2008) ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur J Heart Fail 10: Sharma UC, Pokharel S, van Brakel TJ, van Berlo JH, Cleutjens JP, Schroen B, Andre S, Crijns HJ, Gabius HJ, Maessen J, Pinto YM (2004) Galectin-3 marks activated macrophages in failure-prone hypertrophied hearts and contributes to cardiac dysfunction. Circulation 110: de Boer RA, Yu L, van Veldhuisen DJ (2010) Galectin-3 in cardiac remodeling and heart failure. Curr Heart Fail Rep 7: de Boer RA, Voors AA, Muntendam P, van Gilst WH, van Veldhuisen DJ (2009) Galectin-3: a novel mediator of heart failure development and progression. Eur J Heart Fail 11: Shah RV, Chen-Tournoux AA, Picard MH, van Kimmenade RR, Januzzi JL (2010) Galectin-3, cardiac structure and function, and long-term mortality in patients with acutely decompensated heart failure. Eur J Heart Fail 12: Lok DJ, Van Der Meer P, de la Porte PW, Lipsic E, Van Wijngaarden J, Hillege HL, van Veldhuisen DJ (2010) Prognostic value of galectin-3, a novel marker of fibrosis, in patients with chronic heart failure: data from the DEAL-HF study. Clin Res Cardiol 99: de Boer RA, Lok DJ, Jaarsma T, van der Meer P, Voors AA, Hillege HL, van Veldhuisen DJ (2011) Predictive value of plasma galectin-3 levels in heart failure with reduced and preserved ejection fraction. Ann Med 43: de Boer RA, van Veldhuisen DJ, Gansevoort RT, Muller Kobold AC, van Gilst WH, Hillege HL, Bakker SJ, van der Harst P (2011) The fibrosis marker galectin-3 and outcome in the general population. J Intern Med 11. van Kimmenade RR, Januzzi JL,Jr, Ellinor PT, Sharma UC, Bakker JA, Low AF, Martinez A, Crijns HJ, MacRae CA, Menheere PP, Pinto YM (2006) Utility of amino-terminal pro-brain natriuretic peptide, galectin-3, and apelin for the evaluation of patients with acute heart failure. J Am Coll Cardiol 48: Christenson RH, Duh SH, Wu AH, Smith A, Abel G, defilippi CR, Wang S, Adourian A, Adiletto C, Gardiner P (2010) Multi-center determination of galectin-3 assay performance characteristics: Anatomy of a novel assay for use in heart failure. Clin Biochem 43: Felker GM, Fiuzat M, Shaw LK, Clare R, Whellan DJ, Bettari L, Shirolkar SC, Donahue M, Kitzman DW, Zannad F, Pina IL, O'Connor CM (2012) Galectin-3 in ambulatory patients with heart failure: results from the HF-ACTION study. Circ Heart Fail 5:

125 Chapter Grandin EW, Jarolim P, Murphy SA, Ritterova L, Cannon CP, Braunwald E, Morrow DA (2012) Galectin-3 and the development of heart failure after acute coronary syndrome: pilot experience from PROVE IT-TIMI 22. Clin Chem 58: Lainscak M, Coletta AP, Sherwi N, Cleland JG (2010) Clinical trials update from the Heart Failure Society of America Meeting 2009: FAST, IMPROVE-HF, COACH galectin-3 substudy, HF-ACTION nuclear substudy, DAD-HF, and MARVEL-1. Eur J Heart Fail 12: Hartog JW, Willemsen S, van Veldhuisen DJ, Posma JL, van Wijk LM, Hummel YM, Hillege HL, Voors AA, BENEFICIAL investigators (2011) Effects of alagebrium, an advanced glycation endproduct breaker, on exercise tolerance and cardiac function in patients with chronic heart failure. Eur J Heart Fail 13: Willemsen S, Hartog JW, Hummel YM, Posma JL, van Wijk LM, van Veldhuisen DJ, Voors AA (2010) Effects of alagebrium, an advanced glycation end-product breaker, in patients with chronic heart failure: study design and baseline characteristics of the BENEFICIAL trial. Eur J Heart Fail 12: Willemsen S, Hartog JW, Hummel YM, van Ruijven MH, van der Horst IC, van Veldhuisen DJ, Voors AA (2011) Tissue advanced glycation end products are associated with diastolic function and aerobic exercise capacity in diabetic heart failure patients. Eur J Heart Fail 13: Sheffield LT, Roitman D (1976) Stress testing methodology. Prog Cardiovasc Dis 19: van den Broek SA, van Veldhuisen DJ, de Graeff PA, Landsman ML, Hillege H, Lie KI (1992) Comparison between New York Heart Association classification and peak oxygen consumption in the assessment of functional status and prognosis in patients with mild to moderate chronic congestive heart failure secondary to either ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 70: Hogenhuis J, Voors AA, Jaarsma T, Hoes AW, Hillege HL, Kragten JA, van Veldhuisen DJ (2007) Anaemia and renal dysfunction are independently associated with BNP and NT-proBNP levels in patients with heart failure. Eur J Heart Fail 9: Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D (1999) A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 130: Yang RY, Rabinovich GA, Liu FT (2008) Galectins: structure, function and therapeutic potential. Expert Rev Mol Med 10:e Henderson NC, Mackinnon AC, Farnworth SL, Kipari T, Haslett C, Iredale JP, Liu FT, Hughes J, Sethi T (2008) Galectin-3 expression and secretion links macrophages to the promotion of renal fibrosis. Am J Pathol 172: Smith GL, Lichtman JH, Bracken MB, Shlipak MG, Phillips CO, DiCapua P, Krumholz HM (2006) Renal impairment and outcomes in heart failure: systematic review and meta-analysis. J Am Coll Cardiol 47: Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Charlesworth A, Hampton JR, van Veldhuisen DJ (2000) Renal function, neurohormonal activation, and survival in patients with chronic heart failure. Circulation 102: Smilde TD, Damman K, van der Harst P, Navis G, Westenbrink BD, Voors AA, Boomsma F, van Veldhuisen DJ, Hillege HL (2009) Differential associations between renal function and "modifiable" risk factors in patients with chronic heart failure. Clin Res Cardiol 98: Deepa M. Gopal, Maya Kommineni, Nir Ayalon, Christian Koelbl, Rivka Ayalon, Andreia Biolo, Laura M. Dember, Jill Downing, Deborah A. Siwik, Chang-seng Liang and Wilson (2012) Relationship of Plasma Galectin-3 to Renal Function in Patients With Heart Failure: Effects of Clinical Status, Pathophysiology of Heart Failure, and Presence or Absence of Heart Failure. J Am Heart Assoc 2012, 1: doi: /JAHA

126

127

128 Chapter 7 Summary and Perspectives